JP6706197B2 - Heat exchange system, controller, and method for constructing neural network - Google Patents

Description

The present invention relates to a heat exchange system, a controller, and a method for constructing a neural network.

For example, Patent Document 1 has an air conditioning control characterized by having a neural network that learns operating characteristics due to differences in refrigeration cycles, and a control unit that performs operation control of an air conditioner based on the learning result of the neural network. A device is disclosed.

JP-A-5-10568

In the above air conditioning control device, the temperature fluctuation (physical quantity fluctuation) of the air (fluid) before heat exchange may adversely affect the operation control (heat exchange) of the air conditioner.

INDUSTRIAL APPLICABILITY The present invention relates to a heat exchange system that reduces adverse effects on heat exchange due to changes in physical quantities, a controller that reduces adverse effects on heat exchange due to changes in physical quantities, and a neural network that reduces adverse effects on heat exchange due to changes in physical quantities. The purpose is to provide a construction method of.

(1) In order to achieve the above object, the heat exchange system according to the first aspect of the present invention comprises:
A heat exchange system (for example, a heat exchange system 100) that performs heat exchange using a heat medium (for example, a refrigerant),
An adjusting device (for example, an inverter 102) that controls the heat exchange by adjusting the flow rate of the heat medium,
Feedback control for outputting a manipulated variable to the adjustment device so that the deviation between the first physical quantity (for example, the blown air temperature) that changes according to the flow rate and the target value (for example, the blown air temperature set value) becomes zero. A feedback control unit (for example, the PID control unit 11) to perform,
A value (for example, a current value, a first-order differential value) representing a change in the second physical quantity (for example, intake air temperature) that affects the heat exchange (for example, control of the temperature of the fluid after the heat exchange by the PID control unit 11). And a second-order differential value), and a machine learning unit (for example, a neural network 15 that outputs an operation amount) that controls the adjustment device based on the input value, and the deviation or the operation. A machine learning unit (for example, a neural network 15 that performs learning to reduce the deviation or the manipulated variable to 0) using the amount as a teacher signal to reduce the deviation or the manipulated variable.
Is a heat exchange system including.

(2) In the heat exchange system of (1) above,
The input value includes a first-order differential value and a second-order differential value of the time change of the second physical quantity,
You may do it.

(3) In the heat exchange system according to (1) or (2) above,
The fluctuation of the second physical quantity is a fluctuation of the temperature of the fluid before the heat exchange (for example, the air whose temperature is to be changed by the heat exchange).
You may do it.

(4) In order to achieve the above object, the controller according to the second aspect of the present invention is
A controller (for example, a controller 120) that controls an adjusting device (for example, an inverter 102) that controls the heat exchange by adjusting a flow rate of a heat medium used for heat exchange,
Feedback control for outputting a manipulated variable to the adjustment device so that the deviation between the first physical quantity (for example, the blown air temperature) that changes according to the flow rate and the target value (for example, the blown air temperature set value) becomes zero. A feedback control unit (for example, the PID control unit 11) to perform,
A value (for example, a current value, a first-order differential value) representing a change in the second physical quantity (for example, intake air temperature) that affects the heat exchange (for example, control of the temperature of the fluid after the heat exchange by the PID control unit 11). And a second-order differential value), and a machine learning unit (for example, a neural network 15 that outputs an operation amount) that controls the adjustment device based on the input value, and the deviation or the operation. A machine learning unit that performs learning to reduce the deviation or the operation amount by using the amount as a teaching signal (for example, a neural network 15 that performs learning to set the deviation or the operation amount to 0),
Is a controller equipped with.

(5) In order to achieve the above object, a method for constructing a neural network according to a third aspect of the present invention is
A method for constructing a neural network (for example, a method for producing a neural network 15) that controls an adjusting device (for example, an inverter 102) that controls the heat exchange by adjusting the flow rate of a heat medium used for heat exchange. ,
In feedback control that outputs an operation amount to the adjustment device so that a deviation between a first physical amount (for example, blown air temperature) that changes according to the flow rate and a target value (for example, a blown air temperature set value) becomes zero. A step of causing the neural network (for example, a neural network 15 for learning to set the deviation or the operation amount to 0) to perform learning for reducing the deviation or the operation amount using the deviation or the operation amount as a teacher signal. ,
The neural network has a value (for example, a current value) that represents a change in a second physical quantity (for example, intake air temperature) that affects the heat exchange (for example, control of the temperature of the fluid after the heat exchange by the PID control unit 11). An input value including a first-order differential value and a second-order differential value) is input, and the adjusting device is controlled based on the input value (for example, an operation amount is output),
This is a method of constructing a neural network.

According to the present invention, it is possible to reduce adverse effects on heat exchange due to fluctuations in physical quantities.

It is a block diagram of the heat exchange system which concerns on one embodiment of this invention. It is an example of a block diagram showing a part of the controller of FIG. It is a figure which shows the structural example of a neural network. It is a figure which shows the fluctuation transition (time change) of (A) suction air temperature. (B) is a figure which shows the transition of the air blowing temperature when the variation transition of (A) is given to a heat exchange system, and when a neural network is not used. (C) is a figure which shows the transition of the air blowing temperature when the change transition of (A) is given to a heat exchange system, and when a neural network is used.

[Heat exchange system 100]
The heat exchange system 100 according to the embodiment of the present invention is configured as a direct expansion type cooling device that cools (cools) a room to be cooled.

As shown in FIG. 1, the heat exchange system 100 includes a circulation path L1, a compressor 101, an inverter 102, a condenser 103, a cooling water passage 104, a cooling water flow rate adjusting valve 105, and an electronic expansion valve 106. The fan 107, the evaporator 108, the operation unit 112, the controller 120, the pressure sensors SP1 to SP2, and the temperature sensors ST1 to ST3 are provided.

Further, the heat exchange system 100 is generally configured by an outdoor unit Y and an indoor unit Z. The outdoor unit Y is equipped with a compressor 101, an inverter 102 and the like, and the indoor unit Z is equipped with a blower 107, an evaporator 108 and the like. The indoor unit Z includes a suction port Z1 that sucks air around the indoor unit Z and a blowout port Z2 that blows out the air cooled by the indoor unit Z.

The circulation path L1 is a path for circulating the refrigerant. A compressor 101, a condenser 103, an electronic expansion valve 106, and an evaporator 108 are arranged in the middle of the circulation path L1.

The compressor 101 compresses the refrigerant (vapor) flowing through the circulation path L1 into high temperature and high pressure. The compressor 101 also has a function of circulating a refrigerant. The compressor 101 uses a motor to compress the refrigerant.

The inverter 102 adjusts the output of the compressor 101 (the flow rate of the refrigerant discharged from the compressor 101). The inverter 102 adjusts the output of the compressor 101 by adjusting the rotation speed of the motor of the compressor 101.

The condenser 103 condenses (cools and liquefies) the refrigerant compressed by the compressor 101. A cooling water passage 104 through which cooling water flows passes through the inside of the condenser 103, and the cooling water passing through the cooling water passage 104 condenses the refrigerant.

The cooling water flow rate adjusting valve 105 is arranged in the middle of the cooling water channel 104, and adjusts the flow rate of the cooling water flowing through the cooling water channel 104 by the opening thereof.

A pressure sensor SP1 is provided near the outlet of the condenser 103 (the outlet of the condensed refrigerant). The pressure sensor SP1 converts the pressure of the refrigerant condensed by the condenser 103 (also referred to as condensation pressure) into an electric signal, and supplies the converted electric signal (a signal indicating the condensation pressure (pressure value)) to the controller 120. .. In this way, the pressure sensor SP1 detects the condensation pressure and supplies the detected condensation pressure (pressure value) to the controller 120.

The electronic expansion valve 106 adjusts the flow rate of the refrigerant condensed by the condenser 103 according to the opening degree thereof to rapidly reduce the pressure of the refrigerant. Due to this pressure decrease, the temperature of the refrigerant also decreases.

The blower 107 blows the air in the room to be cooled to the evaporator 108.

The evaporator 108 evaporates the refrigerant whose temperature and pressure have been reduced by the electronic expansion valve 106. The evaporation of the refrigerant by the evaporator 108 removes heat from the air blown by the blower 107. That is, the evaporator 108 exchanges heat between the air that is blown and the refrigerant that evaporates, and cools the air.

The blower 107 and the evaporator 108 are mounted on the indoor unit Z. By the operation of the blower 107, the air in the vicinity of the suction port Z1 of the room to be cooled is sucked into the indoor unit Z from the suction port Z1, the sucked air is cooled by the evaporator 108, and the cooled air is blown out at the outlet Z2. Is blown out from and is supplied to the room to be cooled. As a result, the room to be cooled is cooled.

A temperature sensor ST1 is provided near the suction port Z1. The temperature sensor ST1 converts the temperature near the suction port Z1 (also referred to as suction air temperature) into an electric signal, and supplies the converted electric signal (a signal indicating the suction air temperature (temperature value)) to the controller 120. In this way, the temperature sensor ST1 detects the intake air temperature and supplies the detected intake air temperature (temperature value) to the controller 120. The suction air temperature when the blower 107 is operating is the temperature of the air sucked from the suction port Z1 (air before cooling).

A temperature sensor ST2 is provided near the outlet Z2. The temperature sensor ST2 converts the temperature near the outlet Z2 (also referred to as blowout air temperature) into an electric signal, and supplies the converted electric signal (a signal indicating the blowout air temperature (temperature value)) to the controller 120. In this way, the temperature sensor ST2 detects the blown air temperature and supplies the detected blown air temperature (temperature value) to the controller 120. The temperature of the blown air when the blower 107 is operating can be said to be the temperature of the air (the air blown from the blowout port Z2) after being cooled by the evaporator 108.

A pressure sensor SP2 and a temperature sensor ST3 are provided near the refrigerant outlet of the evaporator 108. The refrigerant outlet is an outlet through which the refrigerant evaporated by the evaporator 108 exits the evaporator. The pressure sensor SP2 converts the pressure of the refrigerant evaporated by the evaporator 108 (also referred to as evaporation pressure) into an electric signal, and supplies the converted electric signal (a signal indicating the evaporation pressure (pressure value)) to the controller 120. In this way, the pressure sensor SP2 detects the evaporation pressure and supplies the detected evaporation pressure (pressure value) to the controller 120. The temperature sensor ST3 converts the temperature of the refrigerant evaporated by the evaporator 108 (also referred to as refrigerant vapor temperature) into an electric signal, and supplies the converted electric signal (a signal indicating the refrigerant vapor temperature (temperature value)) to the controller 120. To do. The temperature sensor ST3 detects the refrigerant vapor temperature and supplies the detected refrigerant vapor temperature (temperature value) to the controller 120.

The refrigerant circulates on the circulation path L1 in the order of compressor 101→condenser 103→electronic expansion valve 106→evaporator 108→compressor 101. Note that an accumulator or the like that stores the refrigerant may be provided in the middle of the circulation path L1.

The operation unit 112 includes a remote controller or the like that receives an operation from the user. The user operates, for example, the operation unit 112 to input desired values for the condensing pressure, the blown air temperature, and the degree of superheat in the heat exchange system 100. The degree of superheat is the degree of superheat of the refrigerant evaporated in the evaporator 108, and is calculated by the refrigerant vapor temperature-saturation temperature (saturation temperature at the evaporation pressure of the refrigerant). The input values are supplied to the controller 120.

The controller 120 is composed of various computers such as PLC (programmable logic controller). For example, the controller 120 is a ROM (Read Only Memory) that stores programs and data, and a nonvolatile storage that can change (including addition and deletion) stored information (programs and data) such as a hard disk and a flash memory. A device (hereinafter, also simply referred to as a non-volatile storage device) and a controller based on a program stored in the ROM or the non-volatile storage device and using data stored in the ROM or the non-volatile storage device. A CPU (Central Processing Unit) that actually executes the process executed by 120 and a RAM (Random Access Memory) that serves as the main memory of the CPU are provided.

The controller 120 stores the above-mentioned respective values supplied from the operation unit 112 in a predetermined storage area of the non-volatile storage device. As a result, each of the above values is set as the set value. The set values thus set are also referred to as the condensing pressure set value, the blown air temperature set value, and the superheat degree set value, respectively. Each of the set values is used as a target value in PID control and the like described later. At least one of the condensation pressure and the degree of superheat may be set as a fixed value by the manufacturer of the heat exchange system 100.

The controller 120 controls the inverter 102, the cooling water flow rate adjusting valve 105, and the electronic expansion valve 106 to control the condensing pressure, the blown air temperature, and the degree of superheat, and bring them close to the respective set values (target values). (Including matching and maintaining). The condensing pressure is controlled by the opening degree (flow rate of cooling water) of the cooling water flow rate adjusting valve 105. The flow rate of the refrigerant (blow-out air temperature) is controlled by the inverter 102. The degree of superheat is controlled by the opening degree (flow rate of the refrigerant) of the electronic expansion valve 106.

The controller 120 supplies an operation amount that can take a numerical value within a predetermined range, for example, to each of the inverter 102, the cooling water flow rate adjustment valve 105, and the electronic expansion valve 106 (hereinafter, also referred to as a control target), Control the controlled object. The operation amount takes a common numerical value range of 0 to 100 [%] for each control target, but the operation based on the same operation amount basically differs depending on the control target.

In the inverter 102, the frequency of the output AC power is controlled by the manipulated variable. The frequency is controlled according to the manipulated variables 0 to 100. The inverter 102 outputs the AC power of the minimum frequency (minimum frequency that the inverter 102 can output) when the manipulated variable is 0 [%], and the maximum frequency (when the inverter 102 is The maximum frequency that can be output) AC power is output. Therefore, when the manipulated variable increases, the frequency increases (the manipulated variable indicates a ratio to the maximum frequency). The frequency of the AC power controls the rotation speed of the motor of the compressor 101. Therefore, the output of the compressor 101 (as a result, the flow rate of the refrigerant output from the compressor 101, the temperature of blown air) is controlled by the manipulated variable (as a result, the heat exchange in the evaporator 108 is controlled). Be done).

The opening degree of each of the cooling water flow rate adjusting valve 105 and the electronic expansion valve 106 is controlled by the operation amount. The operation amount indicates the ratio of the opening degree. Specifically, when the operation amount is 0 [%], the opening degree is 0 (the valve is closed). When the manipulated variable is 100%, the opening is fully open. When the manipulated variable is 20%, the opening is 20% of full opening.

For example, the controller 120 calculates the operation amount by PID (Proportional-Integral-Differential) based on the deviation between the condensation pressure setting value and the condensation pressure (feedback value) detected by the pressure sensor SP2. The operation amount is supplied to the cooling water flow rate adjusting valve 105 (PID control). By such PID control, the opening degree of the cooling water flow rate adjusting valve 105 (in other words, the condensation pressure) is adjusted so that the condensation pressure approaches the condensation pressure set value (target value) (the deviation becomes 0). Controlled.

For example, the controller 120 calculates the degree of superheat derived from the evaporation pressure and the refrigerant vapor temperature detected by the pressure sensor SP2 and the temperature sensor ST3. The controller 120 derives the saturation temperature (saturation temperature at the evaporation pressure) based on the evaporation pressure (may be calculated, or may be calculated by referring to a table that defines the relationship between the evaporation pressure and the saturation temperature). Good)), the value obtained by subtracting the saturation temperature from the refrigerant vapor temperature is acquired as the superheat degree (the superheat degree is fed back). The controller 120 calculates the operation amount by PID based on the deviation between the acquired superheat degree and the superheat degree set value, and supplies the calculated operation amount to the electronic expansion valve 106 (PID control). By such PID control, the opening degree (in other words, the degree of superheat) of the electronic expansion valve 106 is controlled so that the degree of superheat approaches the set value (target value) of the degree of superheat (so that the deviation becomes 0). It

[Controller 120 when controlling blown air temperature (inverter 102)]
As shown in FIG. 2, the controller 120 that controls the inverter 102 includes a PID control unit 11, a fluctuation acquisition unit 13, and a neural network 15. For example, the PLC and the like operate as the PID control unit 11, the fluctuation acquisition unit 13, and the neural network 15, respectively, so that the configuration of FIG. 2 is realized.

The deviation between the blown air temperature set value and the value of the blown air temperature (feedback value) detected by the temperature sensor ST2 is input to the PID control unit 11, and the PID control unit 11 is based on the input deviation. , PID, the operation amount is calculated, and the calculated operation amount is output to the inverter 102. In this way, the PID control unit 11 controls the inverter 102 so that the blown air temperature approaches the blown air temperature set value (target value) (so that the deviation becomes 0).

An electric signal (a signal indicating the intake air temperature, which is an analog signal here) from the temperature sensor ST1 is constantly input to the fluctuation acquisition unit 13. The fluctuation acquisition unit 13 removes noise from the electric signal with a low-pass filter or the like, converts the electric signal into a digital signal by sampling, or the like, and a value indicating the fluctuation of the suction air temperature based on the converted digital signal. Is obtained (here, calculated). In this embodiment, the current value of the intake air temperature (the value of the intake air temperature input this time) and the temporal change of the intake air temperature (the value indicating the past intake air temperature) are used as the values indicating the fluctuations of the intake air temperature. Is stored in a predetermined storage unit, and a first-order differential value (speed of temperature change) and a second-order differential value (acceleration of temperature change) obtained by differentiating are acquired. Note that the first-order differential value is the difference (change amount) between the intake air temperature input last time and the intake air temperature input this time, when Δt (time)=1 (second). The fluctuation of the intake air temperature acquired by the fluctuation acquisition unit 13 is output to the neural network 15. The value indicating the fluctuation of the intake air temperature may be acquired by an analog circuit (for example, a differentiation circuit), converted into a digital value, and input to the neural network 15.

In this embodiment, various sensors such as the temperature sensor ST2 supply the detected physical quantity (temperature or pressure) to the controller 120 in the form of an analog signal (analog/digital conversion is performed by the controller 120). It may be supplied to the controller 120 in the form of a digital signal.

As shown in FIG. 3, the neural network 15 has an input layer, an intermediate layer, and an output layer. The input value to the input layer of the neural network 15 is x _i , the coupling weight between the input layer and the intermediate layer is w _ij , the coupling weight between the intermediate layer and the output layer is w _jk , and the input to the element of the intermediate layer is s _j . Then, the input s _j to the element of the intermediate layer can be expressed by the following formula (1), and the output q _j from the intermediate layer to the output layer can be expressed by the following formulas (2) and (3). The output y _k of 15 can be expressed by the following equation (4). Here, the subscripts i, j, and k are element numbers of the input layer, the intermediate layer, and the output layer, respectively, and n and m are the number of inputs and the number of intermediate elements, respectively. In the following formula, x ₀ =1, s ₀ =1 and q ₀ =1. Also, k=1.

In the input layer, as an input value x _i , a value indicating the fluctuation of the intake air temperature output from the fluctuation acquisition unit 13 (current value of intake air temperature (x ₁ ), first-order differential value (x ₂ ), and The second derivative (x ₃ )) is input. These values are a kind of state quantities that affect the heat exchange by the heat exchange system 100 (for example, various state quantities that affect the cooling characteristics of the heat exchange system 100), but in addition to at least some of these values, Alternatively or in the alternative, other state quantities (state quantities that affect heat exchange by the heat exchange system 100) may be input to the input layer. The state quantity includes a physical quantity, an operation quantity, a set value, and the like, and may be a current value, a differential value thereof with time, or the like. For example, various physical quantities of the refrigerant such as the condensing pressure detected by the pressure sensor SP1, the suction air temperature detected by the temperature sensor ST1, and the superheat degree calculated above may be input as the state quantity. Further, various set values such as the blown air temperature set value, the superheat degree set value, and the condensing pressure set value may be input as the state quantity. Further, as the state quantities, various operation quantities (specifically, the operation quantity of the operation quantity of the past three times input to the electronic expansion valve 106, the operation quantity of the past three times input to the inverter 102, and the like). It is a value indicating a history, and is a change amount of the operation amount (current operation amount-previous operation amount), a first-order differential value (time differential), a second-order differential value (time differential) of a time change of the operation amount of past multiple times. ), etc.) may be input. The operation amount input to the inverter 102 is the sum of the operation amount from the PID control unit 11 and the operation amount from the neural network 15 (details will be described later). The operation amount to 102) may be either one of the operation amounts other than the sum. In the input layer, at least the first-order differential value (x ₂ ) of the present value (x ₁ ), the first-order differential value (x ₂ ) and the second-order differential value (x ₃ ) of the intake air temperature, Alternatively, the second derivative value (x ₃ ) may be input.

The neural network 15 derives the output y ₁ based on the input value. For example, the output y ₁ is calculated by the above equations. The neural network 15 outputs the manipulated variable to the inverter 102 as the output y ₁ .

In such a neural network 15, the deviation input to the PID control unit 11 (see the dotted arrow G1 in FIG. 2) and the operation amount output from the PID control unit 11 (see the dotted arrow G2 in FIG. 2). Machine learning is performed so that any one of them is used as a teacher signal and the deviation or the operation amount is set to zero. For example, the neural network 15 performs machine learning so that the deviation or the operation amount that makes the operation amount 0 can be output. The machine learning is performed by, for example, the error back propagation method, and the weight correction in the error back propagation method is performed by the gradient method. In this way, the result of machine learning is reflected in the neural network 15. The neural network 15 will derive an operation amount based on the result of machine learning. The neural network 15 may stop the learning after, for example, delivering the heat exchange system 100 to the user and then perform the learning to some extent, or may always perform the learning.

After learning, the neural network 15 outputs to the inverter 102 an operation amount that cancels the influence of the fluctuation of the intake air temperature (details will be described later).

The sum of the operation amount from the PID control unit 11 and the operation amount y ₁ from the neural network 15 is input to the inverter 102. In this way, the inverter 102 (in other words, the output (flow rate of the refrigerant) of the compressor 101 and the blown air temperature) is controlled by the neural network 15 and the PID control unit 11.

The neural network 15 may control the blower 107 (the amount of air to be blown) according to the control of the inverter 102. For example, when the inverter 102 or the like is controlled to increase the flow rate of the refrigerant (when the air to be blown is further cooled), the blower 107 may be controlled to increase the air to be blown in order to enhance the cooling capacity. ..

[Effects of this embodiment]
(1) Fluctuations in the intake air temperature adversely affect heat exchange between the refrigerant and the air (control of the blown air temperature by the PID control unit 11). This is because the blown air temperature also depends on the intake air temperature. The above-mentioned adverse effects are, for example, that the blown air temperature does not approach the target temperature, the blown air temperature is not stable, and the like. In this embodiment, the neural network 15 learns the deviation input to the PID control unit 11 or the operation amount output from the PID control unit 11 as a teacher signal (learning to set the deviation or the operation amount to 0). Then, the operation amount that cancels the influence on the heat exchange due to the fluctuation of the intake air temperature is derived and output. Therefore, the adverse effect on the heat exchange due to the fluctuation of the intake air temperature can be reduced (ideally, it can be eliminated). The neural network 15 may perform machine learning to reduce the deviation or the manipulated variable, for example, thereby sequentially deriving the manipulated variable so as to reduce the influence on the heat exchange due to the fluctuation of the intake air temperature. Can be output, and the above adverse effects can be reduced.

In the control by the controller 120, the transition of the blown air temperature by the control with the neural network 15 (control by the PID control unit 11 and the neural network 15) and the control without it (control by the PID control unit 11 only) The difference is shown in FIGS. 4(B) and 4(C). Here, the fluctuation of the intake air temperature shown in FIG. The blown air temperature set value is 20°C. The neural network 15 here has already been learned to some extent. As shown in FIG. 4, when the intake air temperature fluctuates, only with PID control, the blown air temperature is unstable due to the influence of the fluctuation (FIG. 4(B)). On the other hand, with the control of the neural network 15, the blown air temperature is stably changing (FIG. 4(C)). As described above, the control by the neural network 15 cancels the adverse effect on the control of the blown air temperature due to the fluctuation of the intake air temperature, and reduces the adverse effect on the heat exchange.

The adverse effect on the heat exchange due to the fluctuation of the intake air temperature may be reduced by, for example, the feedforward control, but it is difficult to set the parameters of the feedforward control (for example, an error is not allowed). According to the neural network 15, the adverse effect on the heat exchange due to the variation of the intake air temperature can be automatically learned by machine learning, the know-how necessary for setting the parameter becomes unnecessary, and the inconvenience can be solved. Further, the feedforward control cannot cope with the secular change of the cooling characteristic of the heat exchange system 100, but the neural network 15 sequentially learns (learns even after the heat exchange system 100 is installed and operated). , It is possible to cope with the secular change. Furthermore, the neural network 15 can reduce the adverse effect on heat exchange as compared with the feedforward control.

(2) Since the first-order differential value and the second-order differential value of the change over time of the fluctuation are input to the neural network 15 as values representing the fluctuation of the intake air temperature, the neural network 15 changes the fluctuation of the intake air temperature. Can be effectively captured, and effective learning can be performed. Note that either the first-order differential value or the second-order differential value may be input to the neural network 15.

[Modifications, etc.]
The present invention is not limited to the above embodiment, and the above embodiment can be modified as appropriate. Although the example will be shown below as a modified example, at least a part of the modified examples described below can be combined. Further, the configuration disclosed in this specification can be omitted in any configuration as long as there is no technical problem.

(Modification 1)
The blown air temperature (set value) input from the operation unit 112 and set may be room temperature, suction air temperature, or the like. When setting the room temperature, a temperature sensor for detecting the room temperature may be provided, and the value of the room temperature detected by the temperature sensor may be used as a feedback value for PID control or the like. When setting the intake air temperature (the temperature is also close to room temperature), the value of the intake air temperature detected by the temperature sensor ST1 may be used as a feedback value for PID control or the like.

(Modification 2)
Assuming that the evaporation pressure exceeds the upper limit pressure of the refrigerant that can be input to the compressor 101, the refrigerant bypass path (a part of the refrigerant discharged from the compressor 101 is input to the compressor 101 again). Route) may be provided. In this case, an evaporation pressure adjusting valve (a valve that adjusts the flow rate of the refrigerant evaporated in the evaporator 108) is provided between the evaporator 108 and the compressor 101 on the circulation path L1, and when the bypass is used, the evaporation pressure adjusting valve The blown air temperature may be controlled by the opening degree. At this time, the evaporation pressure adjusting valve may be controlled by the neural network 15 and the PID control unit for the evaporation pressure adjusting valve. At this time, the neural network 15 performs learning and the like as in the above-described embodiment so as to reduce the adverse effect on the control of the blown air temperature. It should be noted that a neural network for the evaporation pressure control valve (those learned to reduce the adverse effect like the neural network 15) and the neural network 15 for the inverter 102 may be separately prepared.

(Modification 3)
The neural network 15 controls at least one of the cooling water flow rate adjusting valve 105 and the blower 107 in place of or in addition to at least one of the inverter 102 and the electronic expansion valve 106. Good. For example, the neural network 15 performs machine learning by using the operation amount supplied to the cooling water flow rate adjusting valve 105 from the PID control unit that performs PID control of the cooling water flow rate adjusting valve 105 or the deviation input to the PID control unit as a teacher signal. A numerical value (first-order differential value or the like) representing the intake air temperature is input, and the operation amount is output to the cooling water flow rate adjustment valve 105.

(Modification 4)
Instead of the neural network 15, another control unit for performing machine learning may be provided.

(Modification 5)
The heat exchange system 100 may perform heating (in this case, the circulation direction of the refrigerant is reversed). The heat exchange system 100 only needs to use a heat medium capable of heat exchange. Further, the object to be cooled or heated by the heat exchange system 100 is not limited to air and may be other gas, liquid, etc. as long as it is a fluid. For example, the heat exchange system 100 may be used in a freezer. The heat exchange system is not limited to the direct expansion type cooling system, and may be another heat exchange system. In this case, the objects controlled by the neural network 15 and the PID controller 11 are also changed as appropriate. However, it is desirable that the control target is an adjusting device that adjusts the flow rate of the refrigerant used in the heat exchange system (for example, an inverter that controls the compressor, various valves provided in the flow path of the heat medium). The control target is heat exchange (for example, control of the temperature of the fluid after the heat exchange by the PID control unit 11, the temperature of the fluid after the heat exchange and its fluctuation, or the degree of change of the temperature of the fluid changed by the heat exchange). It is desirable that the adjusting device (for example, an inverter, an electronic expansion valve) that affects () controls (heat exchange). Further, the value input to the neural network 15 is a change in a physical quantity that affects heat exchange (for example, a temperature change of the refrigerant flowing into the evaporator 108, a temperature change of the superheat degree, a pressure of the refrigerant at a predetermined position, etc.). ) (Current value, differential value, etc.).

(Modification 6)
Various values used in the controller 120 (for example, various values input to the PID control unit 11, the neural network 15, etc.) may be appropriately normalized (however, if they are normalized, There is no change).

(Modification 7)
The PID control is feedback control in which a predetermined physical quantity (particularly, a physical quantity that varies according to the flow rate of the refrigerant adjusted by the device controlled by the PID control) is fed back to bring the physical quantity close to a target value. The PID control may be changed to another feedback control. Other feedback control includes PI (Proportional-Integral) control.

(Modification 8)
The physical quantity (temperature or pressure) detected by each of the above sensors may be calculated. For example, the refrigerant vapor temperature may be calculated based on the evaporation pressure and the like.

L1... Circulation path, 11... PID control unit, 13... Fluctuation acquisition unit, 15... Neural network, 100... Heat exchange system, 101... Compressor, 102... Inverter , 103... condenser, 104... cooling water channel, 105... cooling water flow rate adjusting valve, 106... electronic expansion valve, 107... blower, 108... evaporator, 112... Operation part, 120... Controller, SP1-SP2... Pressure sensor, ST1-ST3... Temperature sensor

Claims (5)

A heat exchange system for performing heat exchange using a heat medium,
An adjusting device for controlling the heat exchange by adjusting the flow rate of the heat medium,
A feedback control unit that performs feedback control to output an operation amount to the adjustment device so that a deviation between a first physical amount that changes according to the flow rate and a target value becomes 0;
A machine learning unit that receives an input value including a value that represents a change in the second physical quantity that affects the heat exchange, and that controls the adjustment device based on the input value. A machine learning unit that performs learning to reduce the deviation or the operation amount as a signal,
Heat exchange system with. The input value includes a first-order differential value and a second-order differential value of the time change of the second physical quantity,
The heat exchange system according to claim 1. The change in the second physical quantity is a change in the temperature of the fluid before the heat exchange,
The heat exchange system according to claim 1. A controller for controlling the adjusting device for controlling the heat exchange by adjusting the flow rate of the heat medium used for heat exchange,
A feedback control unit that performs feedback control that outputs an operation amount to the adjustment device so that a deviation between a first physical amount that changes according to the flow rate and a target value becomes 0;
A machine learning unit that receives an input value including a value that represents a change in the second physical quantity that affects the heat exchange, and that controls the adjustment device based on the input value. A machine learning unit that performs learning to reduce the deviation or the operation amount as a signal,
Controller. A method of constructing a neural network for controlling an adjusting device for controlling the heat exchange by adjusting the flow rate of a heat medium used for heat exchange,
The deviation or the operation amount in the feedback control for outputting the operation amount to the adjusting device so that the deviation between the first physical amount that changes according to the flow rate and the target value becomes 0 is the deviation or the operation amount as a teacher signal. A step of causing the neural network to perform learning for reducing
The neural network is input with an input value including a value representing a change in the second physical quantity that affects the heat exchange, and controls the adjusting device based on the input value.
How to build a neural network.