WO2024171341A1

WO2024171341A1 - Inference device, inference method, and method for manufacturing compressor

Info

Publication number: WO2024171341A1
Application number: PCT/JP2023/005248
Authority: WO
Inventors: 航希杉浦; 勇二廣澤; 浩二矢部
Original assignee: 三菱電機株式会社
Priority date: 2023-02-15
Filing date: 2023-02-15
Publication date: 2024-08-22

Abstract

An inference device (10) that infers assembly propriety data relating to whether or not to permit assembly of a compressor (6) for compressing refrigerant includes a data acquisition unit (111) for acquiring input data correlated with the assembly propriety data, and an inference unit (113) for inferring the assembly propriety data, on the basis of the input data acquired by the data acquisition unit (111), by using a trained model (20) for inferring the assembly propriety data, on the basis of the input data.

Description

Inference device, inference method, and method for manufacturing compressor

The present disclosure relates to an inference device, an inference method, and a method for manufacturing a compressor that infers assembly feasibility data regarding whether or not to permit the assembly of a compressor that compresses a refrigerant.

　Conventionally, compressors that compress refrigerants are known. For example, JP 2009-209774 A (Patent Document 1) discloses a compressor that includes a cylinder and a rolling piston that rotates along the inner circumferential surface of the cylinder, and compresses the refrigerant by the rotation of the rolling piston inside the cylinder.

JP 2009-209774 A

According to the compressor disclosed in JP 2009-209774 A (Patent Document 1), by setting the size of the gap between the rolling piston and the cylinder to 0.3% to 0.4% of the displacement volume, it is possible to prevent a decrease in the performance of the compressor. However, because the value of 0.3% to 0.4% of the displacement volume is merely a value derived through experimentation, even if the size of the gap between the rolling piston and the cylinder is set to 0.3% to 0.4% of the displacement volume, it is not necessarily the case that the performance of the compressor actually manufactured will be good.

In addition, when the compressor is assembled by combining multiple parts and the characteristics of the compressor, such as its performance, are checked, and if the characteristics of the compressor do not meet the standards, the assembled compressor must be corrected by hand or must be discarded. In this case, the time required to assemble the compressor and the parts used in assembling the compressor are wasted.

This disclosure has been made to solve the above problems, and aims to provide technology that can eliminate waste in the assembly of compressors.

The inference device according to the present disclosure is an inference device that infers assembly feasibility data regarding whether or not to permit assembly of a compressor that compresses a refrigerant. The inference device includes a data acquisition unit that acquires input data that is correlated with the assembly feasibility data, and an inference unit that infers the assembly feasibility data based on the input data acquired by the data acquisition unit, using a trained model for inferring the assembly feasibility data based on the input data.

The inference method disclosed herein is an inference method in which a computer infers assembly feasibility data regarding whether or not assembly of a compressor that compresses a refrigerant is permitted. The inference method includes, as processing executed by the computer, a step of acquiring input data correlated with the assembly feasibility data, and a step of inferring the assembly feasibility data based on the input data acquired in the acquiring step, using a trained model for inferring the assembly feasibility data based on the input data.

The manufacturing method disclosed herein is a method for manufacturing a compressor by a computer. The manufacturing method includes, as processing executed by the computer, a step of combining a first part with a second part, and a step of inferring assembly feasibility data based on at least one of input data of data indicating individual variations of the first part, data indicating individual variations of the second part, and data indicating individual variations of a combined part formed by combining the first part and the second part, using a trained model for inferring assembly feasibility data regarding whether or not assembly of a compressor is permitted based on the input data.

The manufacturing method according to the present disclosure is a method for manufacturing a compressor by a computer. The manufacturing method includes, as processing executed by the computer, a step of combining a first part with a second part, and a step of inferring assembly feasibility data using a trained model for inferring assembly feasibility data regarding whether or not assembly of a compressor is permitted based on input data, based on input data including at least one of data indicating individual variations of the first part and data indicating individual variations of the second part, data indicating individual variations of a combined part formed by combining the first part and the second part, and data indicating individual variations of a third part to be combined with the combined part.

According to the present disclosure, a trained model can be used to infer whether or not to permit assembly of a compressor based on input data that is correlated with whether or not to permit assembly of a compressor, thereby eliminating waste associated with compressor assembly.

1 is a diagram showing a configuration of a compressor according to a first embodiment; FIG. FIG. 2 is a cross-sectional view of a compression mechanism portion. FIG. 4 is a diagram showing an example of an assembly of a compression mechanism portion. 1 is a diagram showing a configuration of an inference device according to a first embodiment; FIG. 1 is a diagram for explaining an overview of supervised learning. 3 is a diagram for explaining input and output of supervised learning in the inference device according to the first embodiment. FIG. FIG. 2 is a diagram illustrating a configuration of a learning device in a learning phase. FIG. 1 is a diagram illustrating a configuration of a neural network. 11 is a flowchart showing a process executed by a learning device (control unit) in a learning phase. FIG. 13 is a diagram showing the configuration of an inference device in the utilization phase. 13 is a flowchart showing the processing executed by the inference device (control unit) in the utilization phase. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 3 is a diagram for explaining an example of input and output of supervised learning in the inference device according to embodiment 1. FIG. 4 is a flowchart relating to a method of manufacturing a compressor in the inference device according to the first embodiment. 13 is a flowchart relating to a method of manufacturing a compressor in the inference device of embodiment 2. 13 is a flowchart relating to a method of manufacturing a compressor in the inference device of embodiment 3. A figure for explaining a learned model in an inference device related to embodiment 4.

Below, the embodiments of the present disclosure will be described in detail with reference to the drawings. Several embodiments will be described below, but it is planned from the beginning of the application that the configurations described in each embodiment will be appropriately combined. Note that the same or equivalent parts in the drawings will be given the same reference numerals and their description will not be repeated.

Embodiment 1.
[Compressor configuration]
A compressor 6 according to a first embodiment will be described with reference to Fig. 1 to Fig. 4. The compressor 6 can be used in an air conditioner that cools or heats an object to be air-conditioned, such as a room, by circulating a refrigerant through a refrigerant circuit. The compressor 6 may also be used in a refrigeration device that cools an object to be cooled, such as a showcase or a unit cooler, by circulating a refrigerant.

FIG. 1 is a diagram showing the configuration of a compressor 6 according to the first embodiment. In FIG. 1, when the compressor 6 is correctly installed, the horizontal direction of the compressor 6 is defined as the X-axis direction, the vertical direction of the compressor 6 is defined as the Y-axis direction, and the direction perpendicular to the X-axis and Y-axis is defined as the Z-axis direction. In FIG. 1, a longitudinal cross section of the compressor 6 taken along the X-Y plane is shown.

The compressor 6 is a rotary compressor and includes a shell (housing) 60, a compression mechanism 62 for compressing a refrigerant (e.g., a refrigerant gas), an electric motor 61 for supplying power to the compression mechanism 62 for compressing the refrigerant, a shaft 613, a glass terminal 67 for supplying power to the electric motor 61, an accumulator 63 for drawing the refrigerant into the shell 60, and a discharge pipe 66 for discharging the refrigerant compressed by the compression mechanism 62 from inside the shell 60.

The shell 60 houses the electric motor 61, the compression mechanism 62, and the shaft 613. The electric motor 61 is fixed in the shell 60 by press fitting or shrink fitting. The electric motor 61 may have a stator 611 (described later) directly attached to the shell 60 by welding. Inside the shell 60, the compression mechanism 62 is disposed below the electric motor 61. Refrigeration oil is stored at the bottom of the shell 60 to lubricate sliding parts such as the rolling piston 622 (described later). The compression mechanism 62 is connected to the electric motor 61 via the shaft 613.

The accumulator 63 has a suction pipe 64 through which the refrigerant is drawn into the accumulator 63, and a supply pipe 65 that supplies the refrigerant to the compression mechanism 62.

The compressor 6 described above is assembled by joining multiple parts by welding using wax or the like. For example, the shell 60 is constructed by welding shell part 60A arranged on the upper surface of the compressor 6 to shell part 60C arranged on the side of the compressor 6 at welding part W1, and by welding shell part 60B arranged on the lower surface of the compressor 6 to shell part 60C at welding part W2. The shell 60 and the accumulator 63 are welded at welding part W3. The shell 60 and the supply pipe 65 are welded at welding part W4. The shell 60 and the discharge pipe 66 are welded at welding part W5.

FIG. 2 is a diagram showing a cross section of the electric motor 61. FIG. 2 shows a cross section of the electric motor 61 when the electric motor 61 is cut along the X-Z plane at the line A-A' shown in FIG. 1. As shown in FIG. 2, the electric motor 61 comprises a stator 611, a winding 615 wound around the stator 611, and a rotor 612 arranged inside the stator 611. The electric motor 61 is, for example, a PM (Permanent Magnet) motor in which a permanent magnet is provided in the rotor 612.

The stator 611 is formed of an iron core or coil, and includes a stator core 610 with a circular or nearly circular cross section. A central hole 619 with a circular cross section is formed in the center of the stator core 610 for positioning the rotor 612. The rotor 612 can rotate in a direction along the X-Z plane in the central hole 619 formed in the stator core 610.

Furthermore, the stator core 610 has a plurality of slots 614 formed in the circumferential direction. A winding 615 is attached to each of the plurality of slots 614. Power is supplied to the winding 615 via glass terminals 67. Note that the winding 615 may be attached to the stator core 610 using any known winding method, such as distributed winding or concentrated winding, and there are no particular limitations on the method of attaching the winding 615.

The rotor 612 has a circular or nearly circular cross section. The outer diameter of the rotor 612 is smaller than the inner diameter of the stator 611. As a result, the rotor 612 is disposed inside the stator 611 so as to fit into the central hole 619 of the stator core 610 without contacting the stator 611. A shaft hole 616 having a circular cross section for passing the shaft 613 along the Y-axis direction is formed in the center of the rotor 612. In addition, a plurality of air hole portions 617 are formed in the rotor 612 so as to surround the shaft hole portion 616. A plurality of permanent magnets 618 are provided outside the plurality of air hole portions 617. Note that the electric motor 61 is not limited to an IPM (Interior Permanent Magnet) motor in which the permanent magnet 618 is embedded inside the rotor 612, but may be an SPM (Surface Permanent Magnet) motor in which the permanent magnet 618 is attached to the outer circumferential surface of the rotor 612.

FIG. 3 is a diagram showing a cross section of the compression mechanism 62. FIG. 3 shows a cross section of the compression mechanism 62 when the compression mechanism 62 is cut along the X-Z plane at line B-B' shown in FIG. 1. As shown in FIG. 3, the compression mechanism 62 includes a cylinder 621 and a rolling piston 622 arranged inside the cylinder 621.

Cylinder 621 has a circular or nearly circular cross section. A compression chamber 630 having a circular cross section for compressing the refrigerant is formed in the center of cylinder 621 and in which rolling piston 622 is disposed. Rolling piston 622 can rotate in a direction along the X-Z plane in compression chamber 630 formed in cylinder 621.

Furthermore, a back pressure chamber 628 and a vane groove 624 are formed in the cylinder 621. The vane groove 624 connects the compression chamber 630 and the back pressure chamber 628. A long vane 625 is provided in the vane groove 624. In the example of FIG. 3, the vane 625 can slide in the Z-axis direction along the vane groove 624.

The rolling piston 622 has a circular or nearly circular cross section. The rolling piston 622 is attached to the outer periphery of an eccentric shaft portion 626 that has a circular or nearly circular cross section. A shaft hole portion 627 having a circular cross section for passing the shaft 613 along the Y-axis direction is formed in the eccentric shaft portion 626 at a position offset from the center of the rolling piston 622 and the eccentric shaft portion 626. In other words, the shaft 613 is inserted into the rolling piston 622 and the eccentric shaft portion 626 along the Y-axis direction.

The tip of the vane 625 is ideally in contact with a portion of the outer circumferential surface of the rolling piston 622, dividing the compression chamber 630 formed by the inner circumferential surface of the cylinder 621 and the outer circumferential surface of the rolling piston 622 into an intake side and a compression side.

The rolling piston 622 rotates in a direction along the XZ plane in accordance with the rotation of the shaft 613. However, because the shaft 613 is inserted at a position that is off-center of the rolling piston 622, the rolling piston 622 rotates eccentrically along the inner circumferential surface of the cylinder 621, with the off-center position as its axis. When the rolling piston 622 rotates eccentrically within the cylinder 621, part of the outer circumferential surface of the rolling piston 622 ideally comes into close contact with part of the inner circumferential surface of the cylinder 621.

FIG. 4 is a diagram showing an example of the assembly of the compression mechanism 62. As shown in FIG. 4, the vane 625 is attached to the vane groove 624 formed in the cylinder 621, and the rolling piston 622 is attached to the center of the hollow cylindrical cylinder 621, thereby assembling the compression mechanism 62.

As shown in FIG. 1, the compression mechanism 62 further includes an upper frame 623A, a lower frame 623B, an upper muffler 624A, and a lower muffler 624B.

The upper frame 623A and the lower frame 623B support the cylinder 621 and rolling piston 622 of the compression mechanism 62 by sandwiching them from above and below (Y-axis direction). The upper frame 623A supports the cylinder 621 and rolling piston 622 by ideally coming into close contact with the upper parts of the cylinder 621 and rolling piston 622. The lower frame 623B supports the cylinder 621 and rolling piston 622 by ideally coming into close contact with the lower parts of the cylinder 621 and rolling piston 622.

The upper frame 623A and the lower frame 623B allow the shaft 613 to be inserted along the Y-axis direction, and bearings (not shown) support the shaft 613 for rotation in a direction along the X-Z plane. An upper shaft portion 613A constituting part of the long shaft 613 is inserted into the upper frame 623A, and the shaft 613 is rotatably supported by the upper frame 623A at the upper shaft portion 613A. A lower shaft portion 613B constituting part of the long shaft 613 is inserted into the lower frame 623B, and the shaft 613 is rotatably supported by the lower frame 623B at the lower shaft portion 613B.

If the electric motor 61, compression mechanism 62, and shaft 613 are assembled in the shell 60 in the correct position, and the dimensions of each part are accurate, ideally the central axes of each part along the Y-axis direction will coincide. For example, as shown in FIG. 1, the central axis of the upper frame 623A coincides with the central axis of the lower frame 623B. The central axis of the shell 60 also coincides with the central axis of the shaft 613.

[Compressor operation]
In the compressor 6 configured as described above, when a current flows through the windings 615 of the stator 611 by the power supplied from the glass terminal 67, a rotating magnetic field is generated in the stator 611. The permanent magnets 618 act on the rotating magnetic field generated in the stator 611 so as to be attracted to the rotating magnetic field, causing the rotor 612 to rotate. As the rotor 612 rotates, the shaft 613 inserted in the rotor 612 rotates. The rotational force of the shaft 613 is then transmitted to the rolling piston 622, causing the rolling piston 622 to rotate eccentrically along the inner circumferential surface of the cylinder 621.

The refrigerant sucked by the accumulator 63 is supplied to the compression chamber 630 of the compression mechanism 62 via the supply pipe 65. In the compression chamber 630, the rolling piston 622 rotates to compress the refrigerant. As shown in FIG. 3, the compression chamber 630 includes an intake side region where the sucked refrigerant exists, and a compression side region where the compressed refrigerant (hereinafter also referred to as "compressed refrigerant") exists. These intake side and compression side regions are created by the outer circumferential surface of the rolling piston 622 contacting the inner circumferential surface of the cylinder 621 and the tip of the vane 625, respectively. The compressed refrigerant is discharged from the compression side region and rises inside the shell 60 through the upper muffler 624A. Refrigerant oil is mixed into the compressed refrigerant.

The mixture of compressed refrigerant and refrigeration oil is separated into compressed refrigerant and refrigeration oil when passing through the air hole 617 formed in the rotor 612. This makes it possible to prevent the refrigeration oil from flowing into the discharge pipe 66. The compressed refrigerant separated from the refrigeration oil is supplied through the discharge pipe 66 to the high-pressure side of the refrigerant circuit in which the refrigerant circulates.

[Correlation between individual compressor variations and compressor characteristics]
The characteristics of the compressor 6 include the performance of the compressor 6. The performance of the compressor 6 is represented by a coefficient of performance (COP) calculated from the input power of the compressor 6 (input power supplied from the glass terminal 67) and the refrigeration capacity. The coefficient of performance (COP) is characteristic data of the compressor 6 indicating the refrigeration capacity per unit of power (for example, 1 kW). The characteristics of the compressor 6 also include noise data and vibration data of the compressor 6. The noise data of the compressor 6 includes a sound pressure level (for example, in decibels) of the sound (noise) generated from the compressor 6 when the compressor 6 is driven. The vibration data of the compressor 6 includes a vibration level indicating the degree to which the compressor 6 vibrates when the compressor 6 is driven.

Factors that reduce the performance of the compressor 6 include individual variations in each part of the compression mechanism 62. In particular, major factors that affect the performance of the compressor 6 in the compression mechanism 62 include the size of the gap (G1 in FIG. 1) between the rolling piston 622 and the upper frame 623A, the size of the gap (G2 in FIG. 1) between the rolling piston 622 and the lower frame 623B, the size of the gap (G3 in FIG. 3) between the outer circumferential surface of the rolling piston 622 and the inner circumferential surface of the cylinder 621, the size of the gap (G4 in FIG. 3) between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622, the size of the gap (G5 in FIG. 3) between the side surface of the vane 625 in the sliding direction and the vane groove 624 of the cylinder 621, the size of the gap (not shown) between the vane 625 and the upper frame 623A, and the size of the gap (not shown) between the vane 625 and the lower frame 623B. If the gaps between the components of the compression mechanism 62 described above are large, the amount of refrigerant that leaks from the gaps increases, and the compression capacity decreases. Furthermore, if the inner diameter of the rolling piston 622 is larger than the standard, the force of the electric motor 61 is not fully transmitted to the compression mechanism 62, and the compression capacity may decrease. If the inner diameter of the rolling piston 622 is smaller than the standard, the rolling piston 622 and the shaft 613 may come into contact with each other.

If the gaps between the components of the compression mechanism 62 are large, the amount of refrigerant that leaks from the gaps increases, and the refrigeration capacity decreases. The coefficient of performance of the compressor 6 is the refrigeration capacity per unit of power, so if the refrigeration capacity decreases, the coefficient of performance decreases. In other words, if the gaps between the components are large, the performance of the compressor 6 decreases. In particular, the main factors that affect the performance of the compressor 6 include the degree of welding at the welded parts W1 and W2 of the shell 60, the welded part W3 between the shell 60 and the accumulator 63, the welded part W4 between the shell 60 and the supply pipe 65, and the welded part W5 between the shell 60 and the discharge pipe 66. If the weld strength at each weld is low, gaps will occur at the joints between multiple components, and refrigerant may leak.

The noise data and vibration data of the compressor 6 can also fluctuate due to individual variations in each part of the compression mechanism 62. In particular, the main factors that affect the noise data and vibration data of the compressor 6 in the compression mechanism 62 include the concentricity of the upper shaft portion 613A and the lower shaft portion 613B, and the size of the gap (G1 in FIG. 3) between the vane 625 and the vane groove 624.

The concentricity of the upper shaft portion 613A and the lower shaft portion 613B represents the degree of misalignment between the central axis of the upper shaft portion 613A and the central axis of the lower shaft portion 613B, and when the concentricity is 0, the central axis of the upper shaft portion 613A and the central axis of the lower shaft portion 613B completely coincide. In other words, the greater the concentricity of the upper shaft portion 613A and the lower shaft portion 613B, the greater the misalignment of the center of rotation of the shaft 613 above and below the cylinder 621, preventing the transmission of rotational energy from the electric motor 61 to the compression mechanism portion 62 and converting the rotational energy into vibration energy. Therefore, the greater the concentricity of the upper shaft portion 613A and the lower shaft portion 613B, the greater the sound pressure level of the noise of the compressor 6 and the greater the vibration level of the compressor 6, qualitatively. In this way, variations in the concentricity of the upper shaft portion 613A and the lower shaft portion 613B tend to deteriorate the noise and vibration characteristics of the compressor 6.

If the gap between the vane 625 and the vane groove 624 is too large, the vane 625 will be more likely to vibrate in the vane groove 624, and the vane 625 will collide with the vane groove 624, generating vibration energy. In other words, if the gap between the vane 625 and the vane groove 624 is too large, qualitatively the sound pressure level of the noise of the compressor 6 will increase, and the vibration level of the compressor 6 will also increase. Also, if the gap between the vane 625 and the vane groove 624 is too small, vibration energy will be generated due to the frictional force generated between the vane 625 and the vane groove 624. The rotational speed of the motor 61 will pulsate due to the frictional force generated between the vane 625 and the vane groove 624, making the compressor 6 more likely to vibrate and generate noise. In other words, if the gap between the vane 625 and the vane groove 624 is too small, qualitatively the sound pressure level of the noise of the compressor 6 increases, and the vibration level of the compressor 6 increases. In this way, variations in the gap between the vane 625 and the vane groove 624 tend to deteriorate the noise and vibration characteristics of the compressor 6.

In order to prevent the characteristics of the compressor 6 from deteriorating, it is important to ensure that the compression mechanism 62 is airtight. For this reason, each component of the compression mechanism 62 described above is precisely machined and surface treated before assembly. However, there is always individual variation in the dimensions of each component, and gaps may occur due to individual variation in each component. When the compressor 6 is manufactured, the dimensions of each of these components are inspected to see if they are within a predetermined tolerance range, and the compressor 6 is manufactured using components whose dimensions are determined to be within the tolerance range.

The compressor 6 is manufactured by combining the various parts of the compression mechanism 62, and it can be qualitatively understood that the dimensions of each part affect the performance of the compressor 6. However, even if the dimensions of each part of the compression mechanism 62 are within the allowable range, this does not necessarily mean that the dimensions of each gap in the assembled compressor 6 are within the allowable range.

For example, when a part having a dimension close to the lower limit of the tolerance range (e.g., the outer diameter of rolling piston 622) is combined with a part having a dimension close to the upper limit of the tolerance range (e.g., the inner diameter of cylinder 621), the gap between those parts becomes large and the gap dimension may exceed the tolerance range. Also, when a part having a dimension close to the upper limit of the tolerance range (e.g., the outer diameter of rolling piston 622) is combined with a part having a dimension close to the lower limit of the tolerance range (e.g., the inner diameter of cylinder 621), the gap between those parts becomes small and the gap dimension may fall below the tolerance range.

Furthermore, multiple gaps occur when the various parts of the compression mechanism 62 are combined, and these multiple gaps may affect each other. For this reason, even if it is possible to grasp the gaps between two parts to some extent, it is not possible to grasp the individual gaps after the various parts of the compression mechanism 62 are combined, and ultimately, it is difficult to confirm the performance of the compressor 6 as a whole until after the compressor 6 has been manufactured.

Another factor that reduces the characteristics of the compressor 6 is individual variation in each part of the motor 61. In particular, the main factors that affect the performance of the compressor 6 in the motor 61 include the amount of magnetic flux of the rotor 612 that links with the windings 615 and the resistance value of the windings 615. Due to the amount of magnetic flux of the rotor 612 that links with the windings 615, an induced voltage occurs based on the law of electromagnetic induction. The magnitude of the induced voltage is proportional to the amount of magnetic flux of the rotor 612 that links with the windings 615. In other words, the amount of magnetic flux of the rotor 612 that links with the windings 615 corresponds to the induced voltage. In addition, the main factors that affect the noise data and vibration data of the compressor 6 in the motor 61 include the amount of magnetic flux of the rotor 612 that links with the windings 615, the inner diameter roundness of the stator 611, and the amount of eccentricity of the rotor 612.

The amount of magnetic flux of the rotor 612 interlinked with the winding 615 varies mainly depending on the magnetic flux density of the permanent magnet 618 inserted in the rotor 612, the dimensions of the permanent magnet 618, the outer diameter of the rotor 612, and the inner diameter of the stator 611, and the variations in these factors also tend to cause the input power of the compressor 6 (input power supplied from the glass terminal 67) to vary. In general, when the amount of magnetic flux of the rotor 612 decreases, the current flowing to the winding 615 of the motor 61 also increases, causing copper loss to increase, and the driving power supplied to the motor 61 increases. This causes the input power of the motor 61 to vary. Furthermore, in order to control the motor 61, data indicating the amount of magnetic flux of the rotor 612 interlinked with the winding 615 of the stator 611 is input in advance to the control device of the motor 61. However, the amount of magnetic flux input in advance is a representative value that does not reflect the individual variations of the compressor 6. Therefore, the greater the deviation of the actual magnetic flux amount from the representative value input to the control device, the more likely the input to the compressor 6 will fluctuate, making the operation of the motor 61 unstable.

Furthermore, the input power of the compressor 6 varies due to variations in the resistance value of the windings 615. The coefficient of performance of the compressor 6 is the refrigeration capacity per unit of power, so as the input power of the compressor 6 increases, the coefficient of performance decreases. In other words, the performance of the compressor 6 varies depending on the amount of magnetic flux of the rotor 612 that links with the windings 615 and the resistance value of the windings 615.

Furthermore, the greater the variation in the amount of magnetic flux of the rotor 612 interlinked with the windings 615, the more likely the sound pressure level of the noise of the compressor 6 and the more likely the vibration level of the compressor 6 are to fluctuate qualitatively. In this way, the variation in the amount of magnetic flux of the rotor 612 interlinked with the windings 615 of the stator 611 tends to deteriorate the noise and vibration characteristics of the compressor 6.

The inner diameter roundness of the stator 611 indicates whether the circle forming the inner peripheral surface of the stator 611 having a circular cross section is close to a perfect circle, and when the inner diameter roundness is 0, the stator 611 is a perfect circle. As shown in FIG. 2, a gap is generated between the inner peripheral surface of the stator 611 and the outer peripheral surface of the rotating rotor 612, and the size of the gap between the inner peripheral surface of the stator 611 and the outer peripheral surface of the rotor 612 increases or decreases depending on the inner diameter roundness of the stator 611. If the size of this gap fluctuates, the magnetic attraction force acting between the stator 611 and the rotor 612 becomes unstable, and qualitatively, the sound pressure level of the noise of the compressor 6 increases, and the vibration level of the compressor 6 increases. In this way, the variation in the inner diameter roundness of the stator 611 tends to deteriorate the noise and vibration characteristics of the compressor 6.

The eccentricity of the rotor 612 represents the amount of deviation between the rotation axis of the rotor 612 and the ideal position when the rotation axis of the rotor 612 deviates from the ideal position, and when the eccentricity is 0, the rotation axis is located at the ideal position. As described above, the outer peripheral surface of the rolling piston 622 and the inner peripheral surface of the cylinder 621 are ideally in close contact with each other, but depending on the eccentricity of the rotor 612, a gap (G2 in FIG. 3) may be generated between the outer peripheral surface of the rolling piston 622 and the inner peripheral surface of the cylinder 621. Also, as shown in FIG. 3, the tip of the vane 625 and the outer peripheral surface of the rotating rolling piston 622 are ideally in close contact with each other, but depending on the eccentricity of the rotor 612, a gap (G3 in FIG. 3) may be generated between the tip of the vane 625 and the outer peripheral surface of the rolling piston 622. That is, depending on the amount of eccentricity of the rotor 612, the size of the gap between the inner circumferential surface of the stator 611 and the outer circumferential surface of the rotor 612, and the size of the gap between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622, increase or decrease. If the size of these gaps fluctuates, the magnetic attraction force acting between the stator 611 and the rotor 612 becomes unstable, and qualitatively, the sound pressure level of the noise of the compressor 6 increases, and the vibration level of the compressor 6 increases. In this way, the variation in the amount of eccentricity of the rotor 612 tends to deteriorate the noise and vibration characteristics of the compressor 6.

Furthermore, there is an interaction between individual variations in the compression mechanism 62 and individual variations in the electric motor 61. For example, if a large amount of refrigerant leaks through the gaps in the compression mechanism 62, less refrigerant is available for compression, resulting in a smaller refrigeration capacity and a smaller compression torque, which in turn results in a smaller torque generated in the electric motor 61. If the torque of the electric motor 61 is small, this has the effect of reducing the input power to the electric motor 61, and so on. Furthermore, the effect on the electric motor 61 differs depending on the location of the gaps that occur in the compression mechanism 62.

The individual variation of the individual parts, the accuracy of the combination of the multiple parts, and the state of the welding that joins the multiple parts as described above can be affected by the dimensional accuracy of each part processed by the manufacturing device and the working environment of the worker at the manufacturing site. Therefore, the individual variation of the individual parts, the combination of the multiple parts, and the state of the welding that joins the multiple parts can vary depending on the identification information of the manufacturing device, the current generated in the manufacturing device, the voltage generated in the manufacturing device, the time required to manufacture the compressor 6, the welding temperature during the manufacturing of the compressor 6, the welding amount (for example, the amount of wax) in the welding, the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site. For example, the expansion rate of the parts and the work efficiency of the worker can vary depending on the temperature and humidity at the manufacturing site. In addition, when the parts are processed by the manufacturing device, a current or voltage is generated when the blade comes into contact with the parts. The processing state of the parts by the manufacturing device depends on the current or voltage generated in the manufacturing device. Furthermore, if any abnormality occurs in the manufacturing of the compressor 6, the time required to manufacture the compressor 6 may be extended. In this way, the characteristics of the compressor 6 can vary depending on the identification information of the manufacturing device, the current generated in the manufacturing device, the voltage generated in the manufacturing device, the time required to manufacture the compressor 6, the temperature (heat) of the welding during the manufacturing of the compressor 6, the amount of welding (for example, the amount of wax), the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

As such, the characteristics of the compressor are determined by the complex intertwining of the various components in the compression mechanism 62 and the various components in the electric motor 61, so it is difficult to accurately confirm the characteristics of the compressor 6 as a whole until the various components in the compression mechanism 62 and the various components in the electric motor 61 have been combined to manufacture the compressor 6.

In order to check the characteristics of the compressor 6, it is sufficient to inspect the compressor 6 after it has been manufactured. However, it is necessary to stabilize the pressure conditions in the inspection device each time the compressor 6 is inspected, and it would take a lot of time to inspect the performance of all the compressors 6 during the manufacturing process. For this reason, the characteristics of the compressor 6 are usually inspected by sampling inspection. However, as described above, there are individual variations in each component of the compression mechanism 62 and the electric motor 61, and therefore the characteristics of the compressor 6 when multiple components are combined vary from compressor to compressor 6. In other words, even if the dimensions and characteristics of each component of the compression mechanism 62 and the electric motor 61 are guaranteed individually, it is not possible to grasp the synergistic effect when these are combined, and it is difficult to guarantee the characteristics of all the compressors 6 manufactured by sampling inspection.

Furthermore, when the characteristics of the compressor 6 are checked after assembling the compressor 6 by combining multiple parts, if the characteristics of the compressor 6 do not meet the standard values, the assembled compressor 6 must be corrected by manual adjustment or the assembled compressor 6 must be discarded. For example, examples of inspection items during assembly of the compressor 6 include the gap between the stator 611 and the rotor 612 of the electric motor 61, the airtightness or welding state of the compressor 6 to confirm that refrigerant is not leaking from the compressor 6, and noise or vibration during operation of the compressor 6. If the characteristics of the compressor 6 do not meet the standards in these inspections, the compressor 6 must be corrected by manual adjustment or the compressor 6 must be discarded. In this case, the time required to assemble the compressor 6 and the parts used in assembling the compressor 6 are wasted. If it were possible to determine whether or not the compressor 6 can be assembled during the assembly process, taking into account the individual variations of individual parts and the effects of combining multiple parts, workers could avoid the waste of having to rework or discard the assembled compressor 6.

The present disclosure therefore provides a technology that uses AI (Artificial Intelligence) to infer assembly feasibility data based on input data that is correlated with assembly feasibility data regarding whether or not assembly of the compressor 6 is permitted.

[Inference device]
Fig. 5 is a diagram showing the configuration of inference device 10 according to embodiment 1. As shown in Fig. 5, inference device 10 includes, as main functional components, a control unit 11, a storage unit 12, and an input unit 13.

The control unit 11 is a computing entity that executes various processes by executing various programs, and an example of such a computing entity is a computer such as a processor. The processor is, for example, configured with a microcontroller, a CPU (central processing unit), or an MPU (micro-processing unit). The processor has the function of executing various processes by executing programs, but some or all of these functions may be implemented using dedicated hardware circuits such as an ASIC (Application Specific Integrated Circuit), a GPU (Graphics Processing Unit), or an FPGA (Field-Programmable Gate Array). The term "processor" is not limited to a processor in the narrow sense that executes processes using a stored program method such as a CPU or an MPU, but may also include hardwired circuits such as an ASIC, a GPU, or an FPGA. For this reason, the processor may also be interpreted as a processing circuit in which processing is defined in advance by computer-readable code and/or hardwired circuits. The processor may be configured with one chip or multiple chips. Furthermore, the processor and associated processing circuitry may be configured as multiple computers interconnected by wire or wirelessly, such as via a local area network or wireless network. The processor and associated processing circuitry may be configured as a cloud computer that performs remote calculations based on input data and outputs the results of the calculations to other devices in remote locations.

The memory unit 12 is a memory that provides a storage area for temporarily storing program codes or work memory when the control unit 11 executes various programs. The memory unit 12 may be one or more non-transitory computer readable mediums. Examples of the memory unit 12 include volatile memories such as dynamic random access memory (DRAM) and static random access memory (SRAM), or non-volatile memories such as read only memory (ROM) and flash memory. Furthermore, the memory unit 12 may be a storage device that provides a storage area for storing various data required for the control unit 11 to execute various programs. The memory unit 12 may be one or more computer readable storage mediums. Examples of the memory unit 12 include storage devices such as solid state drives (SSDs) and hard disk drives (HDDs).

The input unit 13 is an interface into which input data that is correlated with whether or not the compressor can be assembled is input. For example, the input data input into the input unit 13 includes individual data that indicates individual variations of individual parts, data related to the manufacture of the compressor 6, and data that may arise from the combination of multiple parts.

The control unit 11 includes a data acquisition unit 111, a model generation unit 112, and an inference unit 113.

The data acquisition unit 111 acquires input data input from the input unit 13. For example, the data acquisition unit 111 acquires input data via the input unit 13. The model generation unit 112 generates a trained model 20 (described later) for inferring the assembly feasibility data based on the input data, using learning data 30 (described later) which is a set of the input data and assembly feasibility data relating to whether or not assembly of the compressor 6 is permitted, which is correct answer data corresponding to the input data. The inference unit 113 uses the trained model 20 to infer the assembly feasibility data based on the input data.

[Learning Phase]
6 to 10, an application example of the inference device 10 in the learning phase will be described. As described above, the inference device 10 performs supervised learning using learning data 30, which is a set of input data correlated with assembly feasibility data and assembly feasibility data that is answer data corresponding to the input data. Supervised learning is a method of learning the features of the learning data 30 using a data set of factors and results (labels) and inferring results from the input.

FIG. 6 is a diagram for explaining an overview of supervised learning. As shown in FIG. 6, in the learning phase, the inference device 10 executes a learning program 40 to generate (update) a trained model 20 based on training data 30 including an input 1 and an input 2 (correct answer).

In the utilization phase, the inference device 10 uses the trained model 20 to obtain an output based on the input 1.

FIG. 7 is a diagram for explaining the input and output of supervised learning in the inference device 10 according to the first embodiment. As shown in FIG. 7, in the inference device 10, data correlated with whether the compressor 6 can be assembled is used as the input data for input 1. For example, the input data includes individual data showing individual variations of individual parts, data related to the manufacture of the compressor 6, and data that may arise from the combination of multiple parts. The input data for input 1 can be obtained before assembling the compressor 6 or during the assembly of the compressor 6.

In the inference device 10, assembling feasibility data regarding whether or not the compressor 6 can be assembled is used as input 2, which is the correct answer data. In addition, in the inference device 10, assembling feasibility data regarding whether or not the compressor 6 can be assembled is obtained as output.

FIG. 8 is a diagram showing the configuration of the learning device 110 in the learning phase. The learning device 110 is realized by the control unit 11 of the inference device 10. The learning device 110 is capable of transferring data to and from each of the learning program storage unit 121 and the learned model storage unit 122. The learning program storage unit 121 and the learned model storage unit 122 are realized by the storage unit 12 of the inference device 10.

As shown in FIG. 8, the learning device 110 includes a data acquisition unit 111 and a model generation unit 112. The learning device 110 executes a learning program 40 stored in a learning program storage unit 121 to generate a trained model 20 based on learning data 30 including an input 1 and an input 2 (correct answer).

The data acquisition unit 111 acquires learning data 30 including input 1 and input 2 (correct answer). Specifically, the data acquisition unit 111 acquires, as input 1, input data that is correlated with the assembly feasibility data of the compressor 6. The data acquisition unit 111 acquires, as input 2 (correct answer), the assembly feasibility data of the compressor 6. Specific examples of the input data and the assembly feasibility data will be described later with reference to Figures 13 to 18.

The model generation unit 112 uses the learning data 30 including the input 1 and the input 2 (correct answer) acquired by the data acquisition unit 111 to generate a trained model 20 that infers assembly feasibility data for the compressor 6 based on the input data. The model generation unit 112 stores the generated trained model 20 in the trained model storage unit 122.

FIG. 9 is a diagram showing the configuration of a neural network. The model generation unit 112 generates a trained model 20 by supervised learning, for example, according to a neural network model.

A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. There may be one intermediate layer, or two or more layers.

In Figure 9, a three-layer neural network is shown. In Figure 9, a configuration with three inputs and three outputs is shown. When multiple inputs are input to the input layers X1, X2, and X3, the values multiplied by weights w11 to w16 are input to the intermediate layers Y1 and Y2, and the results are further multiplied by weights w21 to w26 to be output from the output layers Z1, Z2, and Z3. This output result changes depending on the values of the weights w11 to w16 and w21 to w26.

The neural network performs supervised learning based on learning data 30 including input 1 and input 2 (correct answer) acquired by data acquisition unit 111. In other words, the neural network learns by inputting input 1 into the input layer and adjusting the weights so that the result output from the output layer approaches input 2 (correct answer).

The model generation unit 112 generates the trained model 20 by performing supervised learning as described above.

FIG. 10 is a flowchart of the processing executed by the learning device 110 (inference device 10) in the learning phase. Note that FIG. 10 shows the processing executed by the inference device 10 corresponding to the learning device 110. Also, in FIG. 10, "S" is used as an abbreviation for "STEP."

As shown in FIG. 10, the inference device 10 acquires learning data 30 including input 1 and input 2 (correct answer) by the data acquisition unit 111 (S1). Note that the inference device 10 is not limited to acquiring input 1 and input 2 (correct answer) simultaneously, and may acquire input 1 and input 2 (correct answer) at different times.

The inference device 10 generates a trained model 20 by performing supervised learning based on the training data 30 using the model generation unit 112 (S2). The inference device 10 stores the generated trained model 20 in the trained model storage unit 122 (S3) and ends this process.

[Utilization phase]
An application example of the inference device 10 in the utilization phase will be described with reference to Fig. 11 and Fig. 12. Fig. 11 is a diagram showing the configuration of the inference device 10 in the utilization phase. The inference device 10 is capable of transferring data to and from a trained model storage unit 122.

As shown in FIG. 11, the inference device 10 includes a data acquisition unit 111 and an inference unit 113. The inference device 10 uses a trained model 20 to obtain an output based on an input 1.

The data acquisition unit 111 acquires input 1. Specifically, the data acquisition unit 111 acquires input data that is correlated with the assembly feasibility data of the compressor 6 as input 1.

The inference unit 113 uses the trained model 20 to obtain assembly feasibility data of the compressor 6 as output based on the input 1. Specifically, the inference unit 113 reads out the trained model 20 from the trained model storage unit 122. The inference unit 113 uses the trained model 20 to infer assembly feasibility data of the compressor 6 as output based on the input data, which is the input 1 acquired by the data acquisition unit 111.

FIG. 12 is a flowchart of the process executed by the inference device 10 (control unit 11) in the utilization phase. In FIG. 12, "S" is used as an abbreviation for "STEP."

As shown in FIG. 12, the inference device 10 acquires input 1 by the data acquisition unit 111 (S11). The inference device 10 inputs the acquired input 1 to the trained model 20 (S12). The inference device 10 uses the trained model 20 to infer as output the assembly feasibility data of the compressor 6 based on the input data that is correlated with the assembly feasibility data of the compressor 6, which is the input 1 (S13). In this way, the inference device 10 can use the trained model 20 to obtain the assembly feasibility data of the compressor 6 based on the input data that is correlated with the assembly feasibility data of the compressor 6 (for example, individual data that indicates individual variations of the compressor 6). The inference device 10 then ends this process.

[An example of input and output in supervised learning]
An example of input and output of supervised learning in inference device 10 will be described with reference to Figures 13 to 18. Figures 13 to 18 are diagrams for explaining an example of input and output of supervised learning in inference device 10 according to embodiment 1.

In the example of FIG. 13, the size of the gap (hereinafter also referred to as the "air gap") between the stator 611 and rotor 612 of the electric motor 61 is used as the assembly feasibility data for determining whether or not assembly of the compressor 6 is permitted. If the air gap is small, the reliability of the compressor 6 decreases, and the noise or vibration of the compressor 6 worsens. If the air gap is large, the performance of the compressor 6 decreases.

The dimensions of the air gap may be affected by the misalignment of the assembly centers of the stator 611 and the rotor 612, the inner diameter of the stator 611, the outer diameter of the rotor 612, the inclination of the shaft 613, and the state of fixation between the stator 611 and the rotor 612. For example, one factor that may cause the dimensions of the air gap to not meet the standard values is that the shaft 613 is inclined beyond an acceptable range from the central axis of the compressor 6. Even if the inclination of the shaft 613 is within the acceptable range, for example, if the outer diameter of the rotor 612 is larger than the standard and the inner diameter of the stator 611 is smaller than the standard, these dimensional errors may accumulate and cause the dimensions of the air gap to not meet the standard values.

Furthermore, the smaller the outer diameter of the rotor 612, the larger the air gap, and the larger the outer diameter of the rotor 612, the smaller the air gap. The smaller the inner diameter of the stator 611, the smaller the air gap, and the larger the inner diameter of the stator 611, the larger the air gap.

If the inner diameter of each of the upper frame 623A and the lower frame 623B is too small, there is a risk of damaging the shaft 613 when assembling the compressor 6, and the gap between each of the upper frame 623A and the lower frame 623B and the shaft 613 becomes small, making it impossible to form an appropriate oil film. As a result, when the compressor 6 is in operation, each of the upper frame 623A and the lower frame 623B and the shaft 613 will seize up, making it impossible for the shaft 613 to rotate. On the other hand, if the inner diameter of each of the upper frame 623A and the lower frame 623B is too small, the shaft 613 will tilt.

If the minor axis dimension of shaft 613 is short, the bearings in lower frame 623B are reduced, causing shaft 613 to tilt. If the diameter dimension of shaft 613 is large, upper frame 623A and lower frame 623B each come into contact with shaft 613. If the diameter dimension of shaft 613 is small, upper frame 623A and lower frame 623B prevent the central axis from being exposed, causing shaft 613 to tilt.

Therefore, if the inference device 10 trains the learned model 20 by machine learning to infer the air gap dimensions as assembly feasibility data using at least one of data on individual variations of individual components and data that can be generated by combining multiple components as input data, the inference device 10 can infer the air gap dimensions with high accuracy based on the input data. The inference device 10 can check whether the air gap dimensions are appropriate by performing such inference before, during, or after the assembly of the compressor 6. If the inference device 10 determines that the inferred air gap dimensions are appropriate, it proceeds to the next process and continues assembling the compressor 6, and if it determines that the inferred air gap dimensions are not appropriate, it is sufficient to either correct the assembled compressor 6 by hand or to discard the assembled compressor 6.

The data on individual variation of the single component, which is the input data in FIG. 13, includes at least one of the dimensions of the upper frame 623A, the dimensions of the lower frame 623B, the dimensions of the shaft 613, the outer diameter of the rotor 612, and the inner diameter of the stator 611. The dimensions of the upper frame 623A include, for example, the height of the upper frame 623A (length in the Y direction in FIG. 1). The dimensions of the lower frame 623B include, for example, the height of the lower frame 623B (length in the Y direction in FIG. 1). The dimensions of the shaft 613 include the diameter dimension in the cross section of the shaft 613 (X-Z cross section in FIG. 2).

The data that can be generated by combining multiple parts, which is the input data in Figure 13, includes the size of the gap between the shaft 613 and the upper frame 623A, the size of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613, the shrink fit (tightening) between the stator 611 and the rotor 612, and a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B.

In the example of FIG. 14, the airtightness or welding state of the compressor 6 is applied as assembly feasibility data for determining whether or not assembly of the compressor 6 is permitted. The airtightness of the compressor 6 can be confirmed, for example, by detecting the amount of refrigerant or gas bubbles that appear on the liquid surface when the compressor 6 is submerged in liquid, the size of the bubbles, or the frequency at which the bubbles appear. For this reason, the value indicating the airtightness of the compressor 6 can be a value obtained by quantifying or dividing the amount of bubbles, the size of the bubbles, or the frequency at which the bubbles appear. For example, the amount or dimensions of the wax attached to the welded parts W1 to W5 shown in FIG. 1 (for example, the width, thickness, or height of the welded parts) can be applied as a value indicating the welded state of the compressor 6. If the airtightness or welding state of the compressor 6 is poor, refrigerant leakage may occur from the welded parts.

The airtightness or welding state of the compressor 6 may be affected by individual variations in individual parts, the state of welding that joins multiple parts, and the accuracy of the combination of multiple parts. The individual variations in individual parts, the state of welding that joins multiple parts, and the combination of multiple parts may vary depending on the identification information of the manufacturing device (not shown) for manufacturing the compressor 6, the current generated in the manufacturing device, the voltage generated in the manufacturing device, the time required to manufacture the compressor 6, the welding temperature during the manufacturing of the compressor 6, the welding amount in the welding (for example, the amount of wax), the temperature (heat) at the manufacturing site of the compressor 6, and the humidity at the manufacturing site. For example, the expansion rate of the parts and the work efficiency of the worker may vary depending on the temperature and humidity at the manufacturing site. In addition, when a part is processed by the manufacturing device, a current or voltage is generated when a blade comes into contact with the part. The processing state of the part by the manufacturing device depends on the current or voltage generated in the manufacturing device. Furthermore, if any abnormality occurs in the manufacturing of the compressor 6, the time required to manufacture the compressor 6 may be extended. If the amount of wax used during welding (brazing) is small, the welding will be poor, leading to refrigerant leakage from the compressor 6. If the heat temperature during welding is low or the welding time is short, the welding will be poor, leading to refrigerant leakage from the compressor 6. If the discharge current of the manufacturing equipment during welding is small, the welding will be poor, leading to refrigerant leakage from the compressor 6.

Data that may result from the combination of multiple parts includes the dimensions of the shell 60 after welding, and the dimensions of the accumulator 63 after welding. The dimensions of the shell 60 include the width dimension of the shell 60 in its cross section (X-Z cross section). The dimensions of the accumulator 63 include the width dimension of the accumulator 63 in its cross section (X-Z cross section). If the dimensions of the shell 60 after welding or the dimensions of the accumulator 63 after welding are large, the welding will be weak, which may reduce the sealing performance of the compressor 6.

Therefore, the inference device 10 can accurately infer the airtightness or welding state of the compressor 6 based on the input data by training the learned model 20 by machine learning to infer the airtightness or welding state of the compressor 6 as assembly feasibility data using at least one of data on individual variations of individual parts, data on the manufacture of the compressor 6, and data that may be generated by combining multiple parts as input data. The inference device 10 can check whether the airtightness or welding state of the compressor 6 is appropriate by performing such inference before, during, or after the assembly of the compressor 6. Then, if the inference device 10 determines that the inferred airtightness or welding state of the compressor 6 is appropriate, it proceeds to the next process and continues assembling the compressor 6, and if it determines that the inferred airtightness or welding state of the compressor 6 is not appropriate, it is sufficient to either correct the assembled compressor 6 by reworking it or to discard the assembled compressor 6.

The input data in Figure 14, which is data regarding individual variations of individual parts, includes the dimensions of the shell 60 before welding and the dimensions of the accumulator 63 before welding.

The data relating to the manufacture of the compressor 6, which is the input data in FIG. 14, includes at least one of the following: identification information of the manufacturing equipment, the current generated in the manufacturing equipment, the voltage generated in the manufacturing equipment, the noise generated in the manufacturing equipment, the vibration generated in the manufacturing equipment, the time required to manufacture the compressor 6, the welding temperature during the manufacture of the compressor 6, the amount of welding (e.g., the amount of wax), the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

The manufacture of the compressor 6 includes processing a number of parts, such as the stator 611, rotor 612, shaft 613, shell 60, and accumulator 63, and assembling these parts to form the compressor 6. The manufacturing equipment also includes, for example, equipment that uses a blade to process each of the parts, such as the stator 611, rotor 612, and shaft 613. The identification information of the manufacturing equipment includes, for example, a serial number and a management number that identify the manufacturing equipment.

The data that can be generated by combining multiple parts, which is the input data in Figure 14, includes the dimensions of the shell 60 after welding and the dimensions of the accumulator 63 after welding.

In the example of FIG. 15, the noise or vibration of the compressor 6 is applied as assembly feasibility data for determining whether or not assembly of the compressor 6 is permitted. As described above, the individual variations of individual components are cited as factors that cause the noise data and vibration data of the compressor 6 to fluctuate. In addition, the change in the eigenvalue due to welding or assembly is cited as a factor that causes the noise data and vibration data of the compressor 6 to fluctuate. The eigenvalue of the compressor 6 is the natural frequency (resonance frequency) of the compressor 6. The eigenvalue of the compressor 6 can change depending on the strength of the welding. For example, if there is a large amount of wax during welding (brazing), the welding will be strong, and the eigenvalue of the compressor 6 may change. In addition, if the temperature of the heat during welding is high or the welding time is long, distortion may occur in the compression mechanism part 62, and the noise of the compressor 6 will worsen. If the discharge current of the manufacturing device during welding is large, the welding will be strong, and the eigenvalue of the compressor 6 may change.

Furthermore, the accuracy of the combination of multiple parts can be cited as a factor that causes the noise data and vibration data of the compressor 6 to fluctuate. Data that can arise from the combination of multiple parts includes the dimensions of the shell 60 after welding and the dimensions of the accumulator 63 after welding. If the dimensions of the shell 60 after welding or the dimensions of the accumulator 63 after welding are small, the welds will be strong, causing the characteristic values of the compressor 6 to fluctuate and the noise to worsen.

Therefore, if the inference device 10 trains the learned model 20 by machine learning to infer the noise data and vibration data of the compressor 6 as assembly feasibility data using at least one of data on individual variations of individual parts, data on the manufacture of the compressor 6, and data that may be generated by combining multiple parts as input data, the inference device 10 can infer the noise data and vibration data of the compressor 6 with high accuracy based on the input data. The inference device 10 can check whether the noise data and vibration data of the compressor 6 are appropriate by performing such inference before, during, or after the assembly of the compressor 6. Then, if the inference device 10 determines that the inferred noise data and vibration data are appropriate, it proceeds to the next process and continues assembling the compressor 6, and if it determines that the inferred noise data and vibration data are not appropriate, it is sufficient to either correct the assembled compressor 6 by reworking it or to discard the assembled compressor 6.

15, which is input data relating to individual variation of a single component, includes at least one of the dimensions of rolling piston 622, the dimensions of cylinder 621, the dimensions of vane 625, the dimensions of upper frame 623A, the dimensions of lower frame 623B, the outer diameter of stator 611, the inner diameter of stator 611, the width of stator 611, the outer diameter of rotor 612, the inner diameter of rotor 612, the dimensions of shaft 613, the dimensions of shell 60, and the dimensions of accumulator 63. The dimensions of rolling piston 622 include, for example, at least one of the dimensions (outer diameter) of the outer peripheral surface of rolling piston 622 that contacts the inner peripheral surface of cylinder 621 and the tip of vane 625, and the height of rolling piston 622 (length in the Y direction in FIG. 1). The dimensions of the cylinder 621 include, for example, at least one of the dimensions (inner diameter) of the inner peripheral surface of the cylinder 621 in contact with the outer peripheral surface of the rolling piston 622, and the thickness (length in the X direction in FIG. 3) of the vane groove 624 of the cylinder 621 in contact with the side surface in the sliding direction of the vane 625. The dimensions of the vane 625 include, for example, the dimension of the side surface in the sliding direction of the vane 625 in contact with the vane groove 624 of the cylinder 621 (length in the Z direction in FIG. 3), and the width of the vane groove 624 in the direction perpendicular to the sliding direction of the vane 625 (X direction in FIG. 3). The dimensions of the upper frame 623A include, for example, the height of the upper frame 623A (length in the Y direction in FIG. 1). The dimensions of the lower frame 623B include, for example, the height of the lower frame 623B (length in the Y direction in FIG. 1). As shown in FIG. 2, the width dimension of the stator 611 includes the dimension 610A between the outer and inner circumferences of the stator 611 (stator core 610).

The data relating to the manufacture of the compressor 6, which is the input data in FIG. 15, includes at least one of the following: identification information of a manufacturing device (not shown) used to manufacture the compressor 6, the current generated in the manufacturing device, the voltage generated in the manufacturing device, the noise generated in the manufacturing device, the vibration generated in the manufacturing device, the time required to manufacture the compressor 6, the welding temperature during the manufacture of the compressor 6, the amount of welding (e.g., the amount of wax), the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

Data that can be generated by combining multiple parts, which are the input data of Figure 15, include the dimension of the gap between the rolling piston 622 and the upper frame 623A (G1 in Figure 1), the dimension of the gap between the rolling piston 622 and the lower frame 623B (G2 in Figure 1), the dimension of the gap between the outer circumferential surface of the rolling piston 622 and the inner circumferential surface of the cylinder 621 (G3 in Figure 3), the dimension of the gap between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622 (G4 in Figure 3), the dimension of the gap between the side of the vane 625 in the sliding direction and the vane groove 624 of the cylinder 621 (G5 in Figure 3), the dimension of the gap between the vane 625 and At least one of the following is included: the size of the gap with the upper frame 623A (not shown), the size of the gap between the vane 625 and the lower frame 623B (not shown), the size of the gap between the shaft 613 and the upper frame 623A, the size of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613, the shrink fit (tightening) between the stator 611 and the rotor 612, the dimensions of the shell 60 after welding, and the dimensions of the accumulator 63 after welding.

In the example of FIG. 16, the performance (coefficient of performance) of the compressor 6 is applied as assembly feasibility data for determining whether or not assembly of the compressor 6 is permitted. As described above, factors that cause fluctuations in the performance of the compressor 6 include individual variations in individual components. Another factor that causes fluctuations in the performance of the compressor 6 is refrigerant leakage from the compression mechanism 62. The cause of refrigerant leakage from the compression mechanism 62 is gaps that can occur between components when multiple components are combined.

Therefore, if the inference device 10 trains the learned model 20 by machine learning to infer the performance of the compressor 6 as assembly feasibility data using at least one of data related to individual variations of individual components and data that may arise from the combination of multiple components as input data, the inference device 10 can infer the performance of the compressor 6 with high accuracy based on the input data. The inference device 10 can check whether the performance of the compressor 6 is appropriate by performing such inference before, during, or after the assembly of the compressor 6. Then, if the inference device 10 determines that the inferred performance of the compressor 6 is appropriate, it proceeds to the next process and continues assembling the compressor 6, and if it determines that the inferred performance of the compressor 6 is not appropriate, it is sufficient to either correct the assembled compressor 6 by reworking it or to discard the assembled compressor 6.

The input data in FIG. 16, which is data relating to individual variations of individual components, includes at least one of the dimensions of the rolling piston 622, the dimensions of the cylinder 621, the dimensions of the vane 625, the dimensions of the upper frame 623A, and the dimensions of the lower frame 623B.

The data that can be generated by combining multiple parts, which is the input data of FIG. 16, includes at least one of the following: the dimension of the gap between the rolling piston 622 and the upper frame 623A (G1 in FIG. 1), the dimension of the gap between the rolling piston 622 and the lower frame 623B (G2 in FIG. 1), the dimension of the gap between the outer circumferential surface of the rolling piston 622 and the inner circumferential surface of the cylinder 621 (G3 in FIG. 3), the dimension of the gap between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622 (G4 in FIG. 3), the dimension of the gap between the side surface of the vane 625 in the sliding direction and the vane groove 624 of the cylinder 621 (G5 in FIG. 3), the dimension of the gap between the vane 625 and the upper frame 623A (not shown), and the dimension of the gap between the vane 625 and the lower frame 623B (not shown).

In the example of FIG. 17, data that may result from a combination of multiple parts is applied as assembly feasibility data for determining whether or not assembly of compressor 6 is permitted.

Data that can be generated by combining multiple parts includes the dimension of the gap between the rolling piston 622 and the upper frame 623A (G1 in Figure 1), the dimension of the gap between the rolling piston 622 and the lower frame 623B (G2 in Figure 1), the dimension of the gap between the outer circumferential surface of the rolling piston 622 and the inner circumferential surface of the cylinder 621 (G3 in Figure 3), the dimension of the gap between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622 (G4 in Figure 3), the dimension of the gap between the side of the vane 625 in the sliding direction and the vane groove 624 of the cylinder 621 (G5 in Figure 3), the dimension of the gap between the vane 625 and the upper frame 623A (not shown), the dimension of the gap between the vane 625 and the lower frame 623B (not shown), the dimension of the gap between the shaft 613 and the upper frame 623A, the dimension of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613, the shrink fit (tightening) between the stator 611 and the rotor 612, the dimensions of the shell 60 after welding, and at least one of the dimensions of the accumulator 63 after welding.

The above-mentioned data on the various gaps, misalignment of the central axis, and shrink fit (tightening) are factors that affect the air gap of the motor 61, the airtightness or welding condition of the compressor 6, the noise data or vibration data of the compressor 6, and the performance of the compressor 6.

Therefore, if the inference device 10 trains the learned model 20 by machine learning to infer data that may result from a combination of multiple parts as assembly feasibility data using at least one of data on individual variations of individual parts and data on the manufacture of the compressor 6 as input data, the inference device 10 can accurately infer data that may result from a combination of multiple parts based on the input data. The inference device 10 can check whether the data that may result from a combination of multiple parts is appropriate by performing such inference before, during, or after the assembly of the compressor 6. If the inference device 10 determines that the data that may result from the inferred combination of multiple parts is appropriate, it proceeds to the next process and continues assembling the compressor 6, and if it determines that the data that may result from the inferred combination of multiple parts is not appropriate, it can either correct the assembled compressor 6 by reworking it or discard the assembled compressor 6.

The data relating to the individual variation of the individual components, which is the input data in FIG. 17, includes at least one of the following: the dimensions of the rolling piston 622, the dimensions of the cylinder 621, the dimensions of the vane 625, the dimensions of the upper frame 623A, the dimensions of the lower frame 623B, the outer diameter of the stator 611, the inner diameter of the stator 611, the width of the stator 611, the outer diameter of the rotor 612, the inner diameter of the rotor 612, the dimensions of the shaft 613, the dimensions of the shell 60, the dimensions of the accumulator 63, the amount of magnetic flux of the rotor 612 interlinked with the winding 615, and the resistance value of the winding 615.

The data relating to the manufacture of the compressor 6, which is the input data in FIG. 17, includes at least one of the following: identification information of the manufacturing equipment, the current generated in the manufacturing equipment, the voltage generated in the manufacturing equipment, the noise generated in the manufacturing equipment, the vibration generated in the manufacturing equipment, the time required to manufacture the compressor 6, the welding temperature during the manufacture of the compressor 6, the amount of welding (e.g., the amount of wax), the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

In the example of FIG. 18, at least one of the noise data of the compressor 6, the vibration data of the compressor 6, the assembly data related to the assembly state of the compressor 6, whether the compressor 6 can be assembled, and the performance (coefficient of performance) of the compressor 6 is applied as the assembly feasibility data for determining whether the assembly of the compressor 6 is permitted.

The assembly data of the compressor 6 includes at least one of the dimensions of the gap between the stator 611 and rotor 612 of the electric motor 61, a value indicating the airtightness of the compressor 6, a value indicating the welding state of the compressor 6, and a characteristic value of the compressor 6. There are a wide variety of quality check items during assembly of the compressor 6, and multiple parts are combined in a complex manner during assembly of the compressor 6. For this reason, it is difficult for a human being to determine which parts and processing states affect the assembly data related to the assembly state of the compressor 6 described above. However, the inference device 10 can use the trained model 20 to infer the assembly data based on input data that is correlated with the assembly data.

The inference device 10 can accurately infer the assembly feasibility data shown in FIG. 18 based on the input data by training the learned model 20 by machine learning using at least one of data on individual variations of individual components, data on the manufacture of the compressor 6, and data that may arise from the combination of multiple components. The inference device 10 can check whether the assembly feasibility data shown in FIG. 18 is appropriate by performing such inference before, during, or after the assembly of the compressor 6. If the inference device 10 determines that the inferred assembly feasibility data shown in FIG. 18 is appropriate, it proceeds to the next process and continues assembling the compressor 6. If the inferred assembly feasibility data shown in FIG. 18 is determined to be inappropriate, it may be possible to either correct the assembled compressor 6 by reworking it or to discard the assembled compressor 6.

The data relating to the individual variation of the individual components, which is the input data in FIG. 18, includes at least one of the dimensions of the rolling piston 622, the dimensions of the cylinder 621, the dimensions of the vane 625, the dimensions of the upper frame 623A, the dimensions of the lower frame 623B, the outer diameter of the stator 611, the inner diameter of the stator 611, the width of the stator 611, the outer diameter of the rotor 612, the inner diameter of the rotor 612, the dimensions of the shaft 613, the dimensions of the shell 60, the dimensions of the accumulator 63, the amount of magnetic flux of the rotor 612 interlinked with the winding 615, and the resistance value of the winding 615.

The data relating to the manufacture of the compressor 6, which is the input data in FIG. 18, includes at least one of the following: identification information of the manufacturing equipment, the current generated in the manufacturing equipment, the voltage generated in the manufacturing equipment, the noise generated in the manufacturing equipment, the vibration generated in the manufacturing equipment, the time required to manufacture the compressor 6, the welding temperature during the manufacture of the compressor 6, the amount of welding (e.g., the amount of wax), the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

Data that can be generated by combining multiple parts, which are the input data of Figure 18, include the dimension of the gap between the rolling piston 622 and the upper frame 623A (G1 in Figure 1), the dimension of the gap between the rolling piston 622 and the lower frame 623B (G2 in Figure 1), the dimension of the gap between the outer circumferential surface of the rolling piston 622 and the inner circumferential surface of the cylinder 621 (G3 in Figure 3), the dimension of the gap between the tip of the vane 625 and the outer circumferential surface of the rolling piston 622 (G4 in Figure 3), the dimension of the gap between the side of the vane 625 in the sliding direction and the vane groove 624 of the cylinder 621 (G5 in Figure 3), the dimension of the gap between the vane 625 and At least one of the following is included: the size of the gap with the upper frame 623A (not shown), the size of the gap between the vane 625 and the lower frame 623B (not shown), the size of the gap between the shaft 613 and the upper frame 623A, the size of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613, the shrink fit (tightening) between the stator 611 and the rotor 612, the dimensions of the shell 60 after welding, and the dimensions of the accumulator 63 after welding.

The output (assembly feasibility data) of the trained model 20 is not limited to the data exemplified in the above-mentioned Figures 13 to 18, and other data may be applied as long as the data is related to whether or not assembly of the compressor 6 is permitted. Furthermore, the input 1 (input data) of the trained model 20 is not limited to the data exemplified in the above-mentioned Figures 13 to 18, and other data may be applied as long as the data is correlated with the output (assembly feasibility data). Furthermore, the combination of input 1 (input data) and output (assembly feasibility data) may be any combination of data as long as there is a correlation between the two.

[Flowchart for manufacturing a compressor]
With reference to Fig. 19, the inference of assembly feasibility data during the manufacture of the compressor 6 will be described in detail. Fig. 19 is a flowchart relating to the manufacturing method of the compressor 6 in the inference device 10 according to the first embodiment. The flowchart shown in Fig. 19 specifies various processing steps (manufacturing method) for manufacturing the compressor 6 by a computer having the functions of the inference device 10 (control unit 11). In Fig. 19, "S" is used as an abbreviation for "STEP".

As shown in FIG. 19, the inference device 10 combines a first part with a second part (S21). For example, the inference device 10 combines a first part, a cylinder 621, with a second part, a vane 625.

The inference device 10 infers assembly feasibility data using the trained model 20 based on data indicating the individual variation of the first part, data indicating the individual variation of the second part, and data that may result from the combination of the first part and the second part (S22). The data indicating the individual variation of the first part is, for example, the dimension of the vane 625. The data indicating the individual variation of the second part is, for example, the dimension of the cylinder 621. The data that may result from the combination of the first part and the second part is, for example, data indicating the individual variation of the combined part obtained by combining the first part and the second part, and is the dimension of the gap between the vane 625 and the cylinder 621 in the combined part obtained by combining the vane 625 and the cylinder 621.

The inference device 10 uses the dimensions of the vane 625, the dimensions of the cylinder 621, and the dimensions of the gap between the vane 625 and the cylinder 621 as input 1 (input data), and uses the trained model 20 based on the input data to infer assembly feasibility data regarding whether or not assembly of the compressor 6 in the next process is permitted. The assembly feasibility data indicates either that assembly of the compressor 6 is permitted, or that assembly of the compressor 6 is not permitted. For example, if the dimensions of the gap between the vane 625 and the cylinder 621 meet a reference value, the inference device 10 infers assembly feasibility data indicating that assembly of the compressor 6 is permitted, and if the dimensions of the gap between the vane 625 and the cylinder 621 do not meet a reference value, the inference device 10 infers assembly feasibility data indicating that assembly of the compressor 6 is not permitted.

The inference device 10 determines whether or not assembly of the compressor 6 can continue based on the inferred assembly feasibility data (S23).

If the inference device 10 is unable to continue assembling the compressor 6 (NO in S23), for example if the dimension of the gap between the vane 625 and the cylinder 621 does not satisfy the reference value, it performs a process for manually correcting the combined part formed by combining the first part and the second part, or a disposal process for discarding the combined part (S24). For example, the inference device 10 displays an image on a display (not shown) that prompts the worker to manually correct the combined part, or moves the combined part to a disposal route. The inference device 10 then ends this process.

On the other hand, if the inference device 10 is able to continue assembling the compressor 6 (YES in S23), for example if the dimension of the gap between the vane 625 and the cylinder 621 meets the reference value, it combines a third part with the combined parts (S25). This allows the inference device 10 to assemble the compressor 6 by combining the first part, the second part, and the third part. For example, the inference device 10 assembles the compression mechanism 62 that constitutes the compressor 6 by combining the third part, the rolling piston 622, with the combined part that combines the first part, the cylinder 621, with the second part, the vane 625. After that, the inference device 10 ends this process.

Note that in FIG. 19, an example has been described in which the inference device 10 (control unit 11) executes each process, but the inference device 10 (control unit 11) may execute only the inference process of S22, and the remaining processes may be executed by a functional unit other than the inference device 10 (control unit 11) included in the computer.

In addition, FIG. 19 illustrates the process of assembling one combined part (e.g., compression mechanism 62) by combining a first part (e.g., cylinder 621), a second part (e.g., vane 625), and a third part (e.g., rolling piston 622). However, the flowchart shown in FIG. 19 may also be applied to the process of assembling two or more combined parts including other combined parts (e.g., electric motor 61).

As described above, the inference device 10 according to the first embodiment uses the trained model 20 to infer assembly feasibility data based on input data correlated with assembly feasibility data regarding whether assembly of the compressor 6 is permitted. Specifically, the inference device 10 infers assembly feasibility data based on data indicating the individual variation of the first part, data indicating the individual variation of the second part, and data indicating the individual variation of a combined part combining the first part and the second part. This eliminates the need for the inference device 10 to modify the assembled compressor 6 or to discard the assembled compressor 6 after the assembly of the compressor 6 is completed, and can prevent the time required to assemble the compressor 6 and the parts used in the assembly of the compressor 6 from being wasted.

Furthermore, if the assembly of the compressor 6 cannot be continued, the inference device 10 can rework the combined part made up of the first part and the second part. This allows the inference device 10 to reduce the effort and time required for rework compared to detecting that the characteristics of the compressor 6 do not satisfy the reference values during inspection after the assembly of the compressor 6 is completed.

The input data input to the trained model 20 is not limited to including all of data indicating the individual variation of the first part, data indicating the individual variation of the second part, and data indicating the individual variation of the combined part combining the first part and the second part, but may include, for example, only data indicating the individual variation of the first part and data indicating the individual variation of the second part, or only data indicating the individual variation of the combined part combining the first part and the second part. The input data input to the trained model 20 may include at least one of data indicating the individual variation of the first part and data indicating the individual variation of the second part, and data indicating the individual variation of the combined part combining the first part and the second part.

Embodiment 2.
An inference device 10 according to the second embodiment will be described with reference to Fig. 20. Note that, in the following, only the parts of the inference device 10 according to the second embodiment that are different from the inference device 10 according to the first embodiment will be described.

FIG. 20 is a flowchart of a manufacturing method of a compressor 6 in the inference device 10 according to the second embodiment. As shown in FIG. 20, the input data input to the trained model 20 may include data of parts used in the previous process (e.g., dimensions) plus data of parts used in the next process (e.g., dimensions). The flowchart shown in FIG. 20 specifies various processing steps (manufacturing method) for manufacturing a compressor 6 by a computer having the functions of the inference device 10 (control unit 11). Note that in FIG. 20, "S" is used as an abbreviation for "STEP".

As shown in FIG. 20, the inference device 10 combines a first part with a second part (S31). For example, the inference device 10 combines a first part, a cylinder 621, with a second part, a vane 625.

The inference device 10 acquires data indicating the individual variation of the third part to be used in the next process (S32). For example, the inference device 10 acquires the dimensions of the rolling piston 622, which is the third part to be used in the next process.

The inference device 10 infers assembly feasibility data using the trained model 20 based on the data indicating the individual variation of the first part, the data indicating the individual variation of the second part, the data that may arise from the combination of the first part and the second part, and the data indicating the individual variation of the third part (S33). The data indicating the individual variation of the first part is, for example, the dimension of the vane 625. The data indicating the individual variation of the second part is, for example, the dimension of the cylinder 621. The data that may arise from the combination of the first part and the second part is, for example, data indicating the individual variation of the combined part obtained by combining the first part and the second part, and is the dimension of the gap between the vane 625 and the cylinder 621 in the combined part obtained by combining the vane 625 and the cylinder 621.

The inference device 10 uses the dimensions of the vane 625, the dimensions of the cylinder 621, the dimensions of the gap between the vane 625 and the cylinder 621, and the dimensions of the rolling piston 622 as input 1 (input data), and uses the trained model 20 based on the input data to infer assembly feasibility data regarding whether or not assembly of the compressor 6 in the next process is permitted. The assembly feasibility data indicates whether assembly of the compressor 6 is permitted or not permitted. For example, the inference device 10 considers the relationship between the dimensions of the gap between the vane 625 and the cylinder 621 and the dimensions of the rolling piston 622, and infers assembly feasibility data indicating that assembly of the compressor 6 is permitted if these dimensions meet a reference value, and considers the relationship between the dimensions of the gap between the vane 625 and the cylinder 621 and the dimensions of the rolling piston 622, and infers assembly feasibility data indicating that assembly of the compressor 6 is not permitted if these dimensions do not meet a reference value.

The inference device 10 determines whether or not assembly of the compressor 6 can continue based on the inferred assembly feasibility data (S34).

If the inference device 10 is unable to continue assembling the compressor 6 (NO in S34), for example if the dimensions of the gap between the vane 625 and the cylinder 621 and the dimensions of the rolling piston 622 do not meet the reference values, a problem has occurred with the combined part combining the first and second parts, so the inference device 10 performs a process to manually correct the combined part combining the first and second parts, or a disposal process to discard the combined part (S35). For example, the inference device 10 displays an image on a display (not shown) that prompts the worker to manually correct the combined part to match the third part, or moves the combined part to a disposal route. The inference device 10 then ends this process.

On the other hand, if the inference device 10 is able to continue assembling the compressor 6 (YES in S34), for example, if the dimension of the gap between the vane 625 and the cylinder 621 and the dimension of the rolling piston 622 meet the reference values, the inference device 10 combines a third part with the combined parts (S36). This allows the inference device 10 to assemble the compressor 6 by combining the first part, the second part, and the third part. For example, the inference device 10 assembles the compression mechanism unit 62 that constitutes the compressor 6 by combining the third part, the rolling piston 622, with the combined part that combines the first part, the cylinder 621, with the second part, the vane 625. After that, the inference device 10 ends this process.

As described above, the inference device 10 according to the second embodiment infers assembly feasibility data based on data indicating the individual variation of the first part, data indicating the individual variation of the second part, data indicating the individual variation of a combined part formed by combining the first part and the second part, and data indicating the individual variation of the third part. This makes it unnecessary for the inference device 10 to modify the assembled compressor 6 by hand or to discard the assembled compressor 6 after the assembly of the compressor 6 is completed, and it is possible to prevent the time required to assemble the compressor 6 and the parts used in the assembly of the compressor 6 from being wasted.

Furthermore, the inference device 10 infers the assembly feasibility data by considering the relationship between the data indicating the individual variation of the combined part formed by combining the first part and the second part and the data indicating the individual variation of the third part. This allows the inference device 10 to infer with higher accuracy whether or not to permit the assembly of the compressor 6, and can reduce the number of compressors 6 for which assembly is not permitted.

The input data input to the trained model 20 is not limited to including all of the data indicating the individual variation of the first part, the data indicating the individual variation of the second part, and the data indicating the individual variation of the combined part combining the first part and the second part, in addition to the data indicating the individual variation of the third part. For example, the input data input to the trained model 20 may include data indicating the individual variation of the first part and the data indicating the individual variation of the second part, and data indicating the individual variation of the third part, or data indicating the individual variation of the combined part combining the first part and the second part, and data indicating the individual variation of the third part. The input data input to the trained model 20 may include at least one of the data indicating the individual variation of the first part and the data indicating the individual variation of the second part, and the data indicating the individual variation of the combined part combining the first part and the second part, and data indicating the individual variation of the third part to be combined with the combined part.

The data indicating the individual variation of the third part acquired in S32 is not limited to data indicating the individual variation of the third part to be used in the next process (e.g., dimensions), but may be data (e.g., average value, standard deviation) obtained when the variation of the data indicating the individual variation of the multiple third parts for each lot to be used in the next process (e.g., dimensions) is applied to a normal distribution. In other words, even if the third part to be used in the next process has not yet been decided, the inference device 10 may apply data such as the average value for each lot or the standard deviation with respect to a predetermined control value assumed from the data indicating the individual variation of the multiple third parts to be used in the next process as input data for input 1, and use the trained model 20 to infer assembly feasibility data.

Embodiment 3.
An inference device 10 according to the third embodiment will be described with reference to Fig. 21. Note that, in the following, only the parts of the inference device 10 according to the third embodiment that are different from the inference device 10 according to the second embodiment described with reference to Fig. 20 will be described.

FIG. 21 is a flowchart relating to a manufacturing method of a compressor 6 in an inference device 10 according to embodiment 3. When the inference device 10 according to embodiment 3 infers that assembly of the compressor 6 is not permitted, the inference device 10 according to embodiment 3 may change the third part that is to be combined with the combined part of the first part and the second part to another third part. Specifically, as shown in FIG. 21, the inference device 10 according to embodiment 3 determines in S34 that assembly of the compressor 6 cannot be continued based on the assembly feasibility data inferred in S33 (NO in S34), and then determines whether the number of NO determinations in S34 has reached a predetermined number (S37).

If the number of times that the inference device 10 has determined NO in S34 has not reached a predetermined number (NO in S37), the inference device 10 changes the data indicating the individual variation of the third part to be combined with the combined part obtained by combining the first part and the second part (S38). For example, the data indicating the individual variation of the first rolling piston 622 to be combined is changed to data indicating the individual variation of the second rolling piston 622. The inference device 10 then proceeds to the process of S33, and infers assembly feasibility data using the trained model 20 based on the data indicating the individual variation of the first part, the data indicating the individual variation of the second part, the data that may arise from the combination of the first part and the second part, and the data indicating the individual variation of the changed third part (S33). That is, the inference device 10 changes the third part (rolling piston 622) and again infers whether the compressor 6 can be assembled.

On the other hand, if the number of times that the inference device 10 judges NO in S34 reaches a predetermined number (YES in S37), a problem has occurred in the combined part that combines the first part and the second part, so the inference device 10 performs a process to correct the combined part by reworking it, or performs a disposal process to discard the combined part (S35).

As described above, even if the assembly of the compressor 6 cannot be continued due to the combination of a combination part that combines the first part and the second part with a third part, the inference device 10 according to the third embodiment changes the planned third part to another third part without immediately correcting the combination part by rework or discarding the combination part, and infers again whether the assembly of the compressor 6 can be continued. This allows the inference device 10 to prevent the time required to correct the combination part by rework or discard the combination part, and to prevent the combination part from being wasted.

Embodiment 4.
An inference device 10 according to embodiment 4 will be described with reference to Fig. 22. Note that, in the following, only the parts of the inference device 10 according to embodiment 4 that are different from the inference devices 10 according to embodiments 1 to 3 will be described.

FIG. 22 is a diagram for explaining the trained model 20 in the inference device 10 according to embodiment 4. The inference device 10 according to embodiment 4 may include multiple trained models. For example, as shown in FIG. 22, in the inference device 10 according to embodiment 4, the trained model 20 may include a first trained model 201 and a second trained model 202.

The first trained model 201, as shown in FIG. 17, is trained by machine learning to output data that may result from a combination of multiple parts, using at least one of data on individual variations of individual parts and data on the manufacture of the compressor 6 as input 1 (input data). The second trained model 202, as shown in FIG. 18, is trained by machine learning to output assembly feasibility data indicating whether the compressor 6 can be assembled, using data that may result from a combination of multiple parts as input 1 (input data).

The inference device 10 may use the first trained model 201 and the second trained model 202 as described above to infer assembly feasibility data indicating whether the compressor 6 can be assembled, based on input data including at least one of data on individual variations of individual parts and data on the manufacture of the compressor 6. For example, the inference device 10 may use the first trained model 201 to infer data that may result from a combination of multiple parts, based on input data including at least one of data on individual variations of individual parts and data on the manufacture of the compressor 6, and may further use the second trained model 202 to infer assembly feasibility data indicating whether the compressor 6 can be assembled, based on input data including data that may result from a combination of multiple parts inferred using the first trained model 201.

Of course, the inference device 10 may use one trained model 20 to infer assembly feasibility data indicating whether the compressor 6 can be assembled based on input data including at least one of data on individual variations of individual components and data on the manufacture of the compressor 6, rather than using multiple trained models such as the first trained model 201 and the second trained model 202.

<Modification>
The data of each part used in the input 1 may be data obtained by a sampling inspection carried out during the manufacture of the compressor 6. In this way, the more compressors 6 are manufactured and the longer the manufacturing period, the more individual data that can be used for the learning data 30 can be collected.

The individual data on the individual variation of the single component of the input 1 may include data showing individual variation outside the tolerance range of the compressor 6. Specifically, a combination of components that will cause the performance of the compressor 6 to be outside the specification range may be deliberately produced in advance, and the inference device 10 may perform machine learning using learning data including the obtained assembly feasibility data and the individual data of each component used. In this way, by deliberately including individual data showing individual variation outside the tolerance range in the learning data 30, the inference accuracy of the inference device 10 can be improved. During mass production of the compressor 6, the individual variation of the components gathers around the median, so components that deviate from the median are not used. In other words, during mass production of the compressor 6, there are many component dimensions or gaps around the median, so there is also a lot of assembly feasibility data around the median in sampling inspection. Therefore, if an attempt is made to infer the performance of the compressor 6 using component dimensions or gaps close to the upper and lower limits of the specifications, there is a risk that the inference accuracy of the inference device 10 will decrease. Therefore, as described above, by performing machine learning using individual data that indicates individual variations in the compressor 6 that are outside the tolerance range, it is possible to improve the accuracy of inferring assembly feasibility data for compressors 6 that use parts whose part dimensions or gaps are close to the upper and lower limits.

The inference device 10 may be a server device communicatively connected to a control device that controls the compressor 6 via a network, or may be a cloud server. The inference device 10 may acquire input data and assembly feasibility data collected from multiple compressors 6 in the same area as the learning data 30, or may acquire input data and assembly feasibility data collected from multiple compressors 6 in different areas as the learning data 30. In this case, machine learning can be performed taking into account differences in areas by including area information in the learning data 30. The areas may be treated as different areas even if the individual inspection devices that inspect the performance of the compressors 6 are different. After machine learning is performed on a certain compressor 6, machine learning may be performed again on another compressor 6.

The learning algorithm used by the model generation unit 112 of the inference device 10 may be deep learning, which learns to extract the features themselves, or other known methods. For example, the model generation unit 112 may perform machine learning according to genetic programming, functional logic programming, support vector machines, etc.

The inference device 10 described above uses supervised learning, but known learning methods such as unsupervised learning, semi-supervised learning, or reinforcement learning may also be used. For example, when performing unsupervised learning, the inference device 10 may use only the input data of input 1 shown in Figures 13 to 18 as the learning data 30. In the learning phase, the inference device 10 learns the characteristics or trends of the collected input data by clustering the collected input data. Then, in the utilization phase, the inference device 10 uses the learned model 20 to identify the class to which the input data belongs, and outputs the assembly feasibility data of the compressor 6 corresponding to that class as the inference result.

<Summary>
An inference device 10 according to one embodiment infers assembly feasibility data regarding whether or not to permit assembly of a compressor 6 that compresses a refrigerant. The inference device 10 includes a data acquisition unit 111 that acquires input data correlated with the assembly feasibility data, and an inference unit 113 that infers the assembly feasibility data based on the input data acquired by the data acquisition unit 111, using a trained model 20 for inferring the assembly feasibility data based on the input data.

With the above configuration, the inference device 10 can use the trained model 20 to infer assembly feasibility data regarding whether or not assembly of the compressor 6 is permitted based on input data that is correlated with whether or not assembly of the compressor 6 is permitted, thereby eliminating waste related to the assembly of the compressor 6.

The compressor 6 includes a compression mechanism 62 for compressing the refrigerant, an electric motor 61 for supplying power to the compression mechanism 62 for compressing the refrigerant, a shaft 613 for connecting the compression mechanism 62 and the electric motor 61, a shell 60 for housing the compression mechanism 62, the electric motor 61, and the shaft 613, and an accumulator 63 for drawing the refrigerant into the shell 60. The input data indicates individual variations of at least one of the compression mechanism 62, the electric motor 61, the shaft 613, the shell 60, and the accumulator 63.

With the above configuration, the inference device 10 can allow the user to check the assembly feasibility data of the compressor 6 based on the individual variation of at least one of the compression mechanism 62, the electric motor 61, the shaft 613, the shell 60, and the accumulator 63.

The compression mechanism 62 comprises a cylinder 621, a rolling piston 622 that rotates along the inner surface of the cylinder 621 based on power from the electric motor 61, a vane 625 that divides the compression chamber 630 formed by the inner surface of the cylinder 621 and the outer surface of the rolling piston 622 into a suction side and a compression side, an upper frame 623A that supports the rolling piston 622 from above, and a lower frame 623B that supports the rolling piston 622 from below. The input data includes at least one of the following: the dimensions of the rolling piston 622, the dimensions of the cylinder 621, the dimensions of the vane 625, the dimensions of the upper frame 623A, the dimensions of the lower frame 623B, the dimensions of the gap between the rolling piston 622 and the upper frame 623A, the dimensions of the gap between the rolling piston 622 and the lower frame 623B, the dimensions of the gap between the rolling piston 622 and the cylinder 621, the dimensions of the gap between the vane 625 and the rolling piston 622, the dimensions of the gap between the vane 625 and the cylinder 621, the dimensions of the gap between the vane 625 and the upper frame 623A, the dimensions of the gap between the vane 625 and the lower frame 623B, the dimensions of the gap between the shaft 613 and the upper frame 623A, the dimensions of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, and a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613.

According to the above configuration, the inference device 10 can calculate the dimensions of the rolling piston 622, the dimensions of the cylinder 621, the dimensions of the vane 625, the dimensions of the upper frame 623A, the dimensions of the lower frame 623B, the dimensions of the gap between the rolling piston 622 and the upper frame 623A, the dimensions of the gap between the rolling piston 622 and the lower frame 623B, the dimensions of the gap between the rolling piston 622 and the cylinder 621, the dimensions of the gap between the vane 625 and the rolling piston 622, and the dimensions of the gap between the vane 625 and the cylinder 621. The user can confirm the assembly feasibility data for the compressor 6 based on at least one of the following: the dimension, the dimension of the gap between the vane 625 and the upper frame 623A, the dimension of the gap between the vane 625 and the lower frame 623B, the dimension of the gap between the shaft 613 and the upper frame 623A, the dimension of the gap between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, and a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613.

The electric motor 61 includes a stator 611, a winding 615 wound around the stator 611, and a rotor 612 provided inside the stator 611. The input data includes at least one of the following: the outer diameter of the stator 611, the inner diameter of the stator 611, the width of the stator 611, the outer diameter of the rotor 612, the inner diameter of the rotor 612, the amount of magnetic flux of the rotor 612 interlinked with the winding 615, the resistance of the winding 615, and the shrink fit between the stator 611 and the rotor 612.

With the above configuration, the inference device 10 can allow the user to confirm the assembly feasibility data of the compressor 6 based on at least one of the compression mechanism 62, the outer diameter of the stator 611, the inner diameter of the stator 611, the width of the stator 611, the outer diameter of the rotor 612, the inner diameter of the rotor 612, the amount of magnetic flux of the rotor 612 interlinked with the windings 615, the resistance value of the windings 615, and the shrink fit between the stator 611 and the rotor 612.

The input data includes at least one of the following: identification information of a manufacturing device for manufacturing the compressor 6 by processing or combining multiple parts; the current generated in the manufacturing device; the voltage generated in the manufacturing device; the noise generated in the manufacturing device; the vibration generated in the manufacturing device; the time required to manufacture the compressor 6; the welding temperature during the manufacturing of the compressor 6; the welding amount during welding; the temperature at the manufacturing site of the compressor 6; and the humidity at the manufacturing site.

With the above configuration, the inference device 10 can allow the user to confirm the assembly feasibility data for the compressor 6 based on at least one of the identification information of the manufacturing equipment, the current generated in the manufacturing equipment, the voltage generated in the manufacturing equipment, the noise generated in the manufacturing equipment, the vibration generated in the manufacturing equipment, the time required to manufacture the compressor 6, the welding temperature during the manufacturing of the compressor 6, the welding amount during welding, the temperature at the manufacturing site of the compressor 6, and the humidity at the manufacturing site.

The assembly feasibility data includes at least one of the following: the size of the gap between the stator 611 and rotor 612 of the electric motor 61 of the compressor 6, a value indicating the airtightness of the compressor 6, a value indicating the welding condition of the compressor 6, noise data indicating the noise of the compressor 6, vibration data indicating the vibration of the compressor 6, and the performance of the compressor 6.

With the above configuration, the inference device 10 can allow the user to confirm at least one of the following based on the input data of the compressor 6: the dimension of the gap between the stator 611 and the rotor 612, a value indicating the airtightness of the compressor 6, a value indicating the welding condition of the compressor 6, noise data indicating the noise of the compressor 6, vibration data indicating the vibration of the compressor 6, and the performance of the compressor 6.

Assembly feasibility data includes combination data that is generated by combining multiple parts.

With the above configuration, the inference device 10 can allow the user to check combination data that is generated by combining multiple parts based on the input data of the compressor 6.

The compressor 6 includes a compression mechanism 62 for compressing the refrigerant, an electric motor 61 for supplying power to the compression mechanism 62 for compressing the refrigerant, a shaft 613 for connecting the compression mechanism 62 and the electric motor 61, a shell 60 for housing the compression mechanism 62, the electric motor 61, and the shaft 613, and an accumulator 63 for drawing the refrigerant into the shell 60. The compression mechanism 62 includes a cylinder 621, a rolling piston 622 that rotates along the inner circumferential surface of the cylinder 621 based on power from the electric motor 61, a vane 625 that divides the compression chamber 630 formed by the inner circumferential surface of the cylinder 621 and the outer circumferential surface of the rolling piston 622 into a suction side and a compression side, an upper frame 623A that supports the rolling piston 622 from above, and a lower frame 623B that supports the rolling piston 622 from below. The electric motor 61 includes a stator 611 , a winding 615 wound around the stator 611 , and a rotor 612 provided inside the stator 611 . The combination data includes at least one of the following: the gap size between the rolling piston 622 and the upper frame 623A, the gap size between the rolling piston 622 and the lower frame 623B, the gap size between the rolling piston 622 and the cylinder 621, the gap size between the vane 625 and the rolling piston 622, the gap size between the vane 625 and the cylinder 621, the gap size between the vane 625 and the upper frame 623A, the gap size between the vane 625 and the lower frame 623B, the gap size between the shaft 613 and the upper frame 623A, the gap size between the shaft 613 and the lower frame 623B, a value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B, a value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613, the shrink fit between the stator 611 and the rotor 612, the dimensions of the shell 60 after welding, and the dimensions of the accumulator 63 after welding.

With the above configuration, the inference device 10, based on the input data of the compressor 6, determines the size of the gap between the rolling piston 622 and the upper frame 623A, the size of the gap between the rolling piston 622 and the lower frame 623B, the size of the gap between the rolling piston 622 and the cylinder 621, the size of the gap between the vane 625 and the rolling piston 622, the size of the gap between the vane 625 and the cylinder 621, the size of the gap between the vane 625 and the upper frame 623A, the size of the gap between the vane 625 and the lower frame 623B, The user can confirm at least one of the following: the size of the gap between the stator 611 and the rotor 612; the size of the gap between the shaft 613 and the upper frame 623A; the size of the gap between the shaft 613 and the lower frame 623B; the value indicating the deviation between the central axis of the upper frame 623A and the central axis of the lower frame 623B; the value indicating the deviation between the central axis of the shell 60 and the central axis of the shaft 613; the shrink fit between the stator 611 and the rotor 612; the dimensions of the shell 60 after welding; and the dimensions of the accumulator 63 after welding.

The trained model 20 includes a first trained model 201 and a second trained model 202. The inference unit 113 uses the first trained model 201 to infer combination data based on the input data acquired by the data acquisition unit 111, and uses the second trained model 202 to infer assembly feasibility data.

With the above configuration, the inference device 10 can use the first trained model 201 and the second trained model 202 to allow the user to confirm the assembly feasibility data of the compressor 6 based on the input data of the compressor 6.

The inference unit 113 uses the trained model 20 to infer whether or not assembly of the compressor is permitted based on the input data acquired by the data acquisition unit 111 as assembly feasibility data.

With the above configuration, the inference device 10 can prompt the user to confirm whether or not to allow assembly of the compressor 6 based on the input data of the compressor 6.

The inference method according to one embodiment is an inference method in which a computer infers assembly feasibility data regarding whether or not assembly of a compressor 6 that compresses a refrigerant is permitted. The inference method includes, as processing executed by the computer, a step (S11) of acquiring input data correlated with the assembly feasibility data, and a step (S13) of inferring the assembly feasibility data based on the input data acquired in the acquiring step, using a trained model 20 for inferring the assembly feasibility data based on the input data.

With the above configuration, the computer can use the trained model 20 to infer assembly feasibility data regarding whether or not assembly of the compressor 6 is permitted based on input data that is correlated with whether or not assembly of the compressor 6 is permitted, thereby eliminating waste associated with the assembly of the compressor 6.

The manufacturing method of the compressor 6 according to one embodiment includes, as a process executed by a computer, a step (S21) of combining a first part with a second part, and a step (S22) of inferring assembly feasibility data based on at least one of input data of data indicating individual variations of the first part, data indicating individual variations of the second part, and data indicating individual variations of a combined part formed by combining the first part and the second part, using a trained model 20 for inferring assembly feasibility data regarding whether or not assembly of the compressor 6 is permitted based on the input data.

With the above configuration, during the manufacturing process of the compressor 6, the computer can use the trained model 20 to infer assembly feasibility data regarding whether or not assembly of the compressor 6 is permitted based on data indicating individual variations of the first and second parts that constitute the compressor 6, thereby eliminating waste associated with the assembly of the compressor 6.

The manufacturing method of a compressor 6 according to one embodiment includes, as a process executed by a computer, a step (S31) of combining a first part with a second part, and a step (S33) of inferring assembly feasibility data based on input data including at least one of data indicating individual variations of the first part and data indicating individual variations of the second part, data indicating individual variations of a combined part formed by combining the first part and the second part, and data indicating individual variations of a third part to be combined with the combined part, using a trained model for inferring assembly feasibility data regarding whether or not assembly of a compressor is permitted based on the input data.

With the above configuration, during the manufacturing process of the compressor 6, the computer can use the trained model 20 to infer assembly feasibility data regarding whether or not to permit assembly of the compressor 6 based on data indicating the individual variations of the first and second parts that make up the compressor 6, as well as data indicating the individual variations of the third part that is planned to be combined with the first and second parts, thereby eliminating waste associated with the assembly of the compressor 6.

　Further includes a step (S38) of changing the third part to be combined with the combination part to another third part when the inference step infers that assembly of the compressor 6 is not permitted.

With the above configuration, if the computer infers that assembly of the compressor 6 is not permitted during the manufacturing process of the compressor 6, it can change the third part that is to be combined with the combined part to another third part, thereby preventing the time required to rework or discard the combined part and preventing the combined part from being wasted.

The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The scope of the present disclosure is indicated by the claims, not by the description of the embodiments above, and is intended to include all modifications within the meaning and scope of the claims.

6 Compressor, 10 Inference device, 11 Control unit, 12 Memory unit, 13 Input unit, 20 Learned model, 30 Learning data, 40 Learning program, 60 Shell, 60A, 60B, 60C Shell parts, 61 Electric motor, 62 Compression mechanism unit, 63 Accumulator, 64 Suction pipe, 65 Supply pipe, 66 Discharge pipe, 67 Glass terminal, 110 Learning device, 111 Data acquisition unit, 112 Model generation unit, 113 Inference unit, 121 Learning program memory unit, 122 Learned model memory unit, 201 First learned model, 202 Second 2 trained model, 610 stator core, 611 stator, 612 rotor, 613 shaft, 613A upper shaft section, 613B lower shaft section, 614 slot, 615 winding, 616, 627 shaft hole section, 617 air hole section, 618 permanent magnet, 619 central hole section, 621 cylinder, 622 rolling piston, 623A upper frame, 623B lower frame, 624 vane groove, 624A upper muffler, 624B lower muffler, 625 vane, 626 eccentric shaft section, 628 back pressure chamber, 630 compression chamber.

Claims

An inference device that infers assembly feasibility data regarding whether or not assembly of a compressor that compresses a refrigerant is permitted,
A data acquisition unit that acquires input data correlated with the assembly feasibility data;
an inference unit that infers the assembly possible/prohibited data based on the input data acquired by the data acquisition unit, using a trained model for inferring the assembly possible/prohibited data based on the input data.
The compressor includes a compression mechanism for compressing the refrigerant, an electric motor for supplying power to the compression mechanism for compressing the refrigerant, a shaft for connecting the compression mechanism and the electric motor, a shell for accommodating the compression mechanism, the electric motor, and the shaft, and an accumulator for drawing the refrigerant into the shell,
The inference device according to claim 1 , wherein the input data indicates individual variations of at least one of the compression mechanism, the electric motor, the shaft, the shell, and the accumulator.
the compression mechanism includes a cylinder, a rolling piston that rotates along an inner peripheral surface of the cylinder based on the power from the electric motor, a vane that divides a compression chamber formed by the inner peripheral surface of the cylinder and the outer peripheral surface of the rolling piston into a suction side and a compression side, an upper frame that supports the rolling piston from above, and a lower frame that supports the rolling piston from below,
3. The inference apparatus of claim 2, wherein the input data includes at least one of a dimension of the rolling piston, a dimension of the cylinder, a dimension of the vane, a dimension of the upper frame, a dimension of the lower frame, a gap dimension between the rolling piston and the upper frame, a gap dimension between the rolling piston and the lower frame, a gap dimension between the rolling piston and the cylinder, a gap dimension between the vane and the rolling piston, a gap dimension between the vane and the cylinder, a gap dimension between the vane and the upper frame, a gap dimension between the vane and the lower frame, a gap dimension between the shaft and the upper frame, a gap dimension between the shaft and the lower frame, a value indicating a deviation between a central axis of the upper frame and a central axis of the lower frame, and a value indicating a deviation between a central axis of the shell and a central axis of the shaft.
The electric motor includes a stator, a winding wound around the stator, and a rotor provided inside the stator,
4. The inference device according to claim 2 or claim 3, wherein the input data includes at least one of an outer diameter dimension of the stator, an inner diameter dimension of the stator, a width dimension of the stator, an outer diameter dimension of the rotor, an inner diameter dimension of the rotor, an amount of magnetic flux of the rotor interlinked with the winding, a resistance value of the winding, and a shrink fit between the stator and the rotor.
The inference device according to any one of claims 1 to 4, wherein the input data includes at least one of the following: identification information of a manufacturing device for manufacturing the compressor by processing or combining a plurality of parts; a current generated in the manufacturing device; a voltage generated in the manufacturing device; a noise generated in the manufacturing device; a vibration generated in the manufacturing device; a time required to manufacture the compressor; a welding temperature during the manufacturing of the compressor; a welding amount in the welding; a temperature at the manufacturing site of the compressor; and a humidity at the manufacturing site.
The inference device according to any one of claims 1 to 5, wherein the assembly feasibility data includes at least one of the following: the size of the gap between the stator and rotor of the electric motor included in the compressor, a value indicating the airtightness of the compressor, a value indicating the welding condition of the compressor, noise data indicating the noise of the compressor, vibration data indicating the vibration of the compressor, and performance of the compressor.
The inference device according to any one of claims 1 to 6, wherein the assembly feasibility data includes combination data resulting from combining multiple parts.
The compressor includes a compression mechanism for compressing the refrigerant, an electric motor for supplying power to the compression mechanism for compressing the refrigerant, a shaft for connecting the compression mechanism and the electric motor, a shell for accommodating the compression mechanism, the electric motor, and the shaft, and an accumulator for drawing the refrigerant into the shell,
the compression mechanism includes a cylinder, a rolling piston that rotates along an inner peripheral surface of the cylinder based on the power from the electric motor, a vane that divides a compression chamber formed by the inner peripheral surface of the cylinder and the outer peripheral surface of the rolling piston into a suction side and a compression side, an upper frame that supports the rolling piston from above, and a lower frame that supports the rolling piston from below,
The electric motor includes a stator, a winding wound around the stator, and a rotor provided inside the stator,
8. The inference device of claim 7, wherein the combination data includes at least one of a gap dimension between the rolling piston and the upper frame, a gap dimension between the rolling piston and the lower frame, a gap dimension between the rolling piston and the cylinder, a gap dimension between the vane and the rolling piston, a gap dimension between the vane and the cylinder, a gap dimension between the vane and the upper frame, a gap dimension between the vane and the lower frame, a gap dimension between the shaft and the upper frame, a gap dimension between the shaft and the lower frame, a value indicating a deviation between a central axis of the upper frame and a central axis of the lower frame, and a value indicating a deviation between a central axis of the shell and a central axis of the shaft, a shrink fit between the stator and the rotor, a dimension of the shell after welding, and a dimension of the accumulator after welding.
The trained model includes a first trained model and a second trained model,
The inference unit is
Using the first trained model, inferring the combination data based on the input data acquired by the data acquisition unit;
The inference device according to claim 7 or claim 8, wherein the assembly feasibility data is inferred based on the combination data by using the second trained model.
The inference device according to any one of claims 1 to 9, wherein the inference unit uses the trained model to infer whether or not assembly of the compressor is permitted as the assembly feasibility data based on the input data acquired by the data acquisition unit.
An inference method for inferring assembly feasibility data regarding whether or not assembly of a compressor that compresses a refrigerant is permitted, by a computer, comprising:
The process executed by the computer is
acquiring input data correlated with the assembly feasibility data;
and inferring the assembly feasibility data based on the input data acquired by the acquiring step, using a trained model for inferring the assembly feasibility data based on the input data.
1. A computer-implemented method for manufacturing a compressor, comprising:
The process executed by the computer is
combining a second part with the first part;
inferring assembly feasibility data regarding whether or not assembly of the compressor is permitted based on at least one input data of data indicating individual variations of the first part, data indicating individual variations of the second part, and data indicating individual variations of a combined part obtained by combining the first part and the second part, using a trained model for inferring assembly feasibility data regarding whether or not assembly of the compressor is permitted based on the input data.
1. A computer-implemented method for manufacturing a compressor, comprising:
The process executed by the computer is
combining a second part with the first part;
and inferring assembly feasibility data regarding whether or not assembly of the compressor is permitted based on input data including at least one of data indicating individual variations of the first part and data indicating individual variations of the second part, data indicating individual variations of a combined part formed by combining the first part and the second part, and data indicating individual variations of a third part to be combined with the combined part, using a trained model for inferring assembly feasibility data regarding whether or not assembly of the compressor is permitted based on the input data.
The method for manufacturing a compressor according to claim 13, further comprising the step of changing the third part to be combined with the combination part to another third part when the inference step infers that assembly of the compressor is not permitted.