JP7574969B1 - ROLLING EMULATION DEVICE, ROLLING EMULATION PROGRAM, AND ROLLING CONTROL SYSTEM - Google Patents

Abstract

The rolling emulation device executes the following processes: a process of acquiring control signals used to control multiple rolling mills from a controller that controls the multiple rolling mills via a communication interface; a process of emulating the transportation of the rolled material in the rolling line by inputting the control signals into a rolling conveying model that simulates the transportation of the rolled material in the rolling line; a process of generating simulated sensor signals that simulate sensor signals output from multiple sensors provided in the rolling line based on the emulation results by the rolling conveying model; a process of acquiring actual sensor signals output from the multiple sensors; a learning process that reflects a model correction value in the rolling conveying model to reduce the difference between the simulated sensor signal and the actual sensor signal; and an abnormality determination process that determines that an abnormality has occurred in one of the multiple rolling mills if the difference is equal to or greater than a first threshold value.

Description

This disclosure relates to technology for simulating a rolling line.

Patent Document 1 discloses a rolling simulation device that simulates the state of a rolling line. Specifically, the rolling simulation device uses a model formula that simulates each process in the rolling line to calculate a target value for rolling metal material and a predicted state value of the rolled material when the process is changed. Furthermore, the rolling simulation device updates the parameters of the model formula based on a comparison value obtained by comparing the predicted state value of the rolled material with the control target value of the actuator used to drive the rolling line. This updates parameters that change over time, improving the prediction accuracy of the model formula.

International Publication No. 2016/038705

The rolling state of the rolling line is reflected in the actual measured value of the sensor installed in the rolling line. One of the reasons why the actual measured value of the rolling state becomes inappropriate is thought to be that an abnormality has occurred in one of the rolling mills constituting the rolling line. In that case, it is sufficient to notify an operator that an abnormality has occurred in the rolling mill without changing the model simulating the rolling line. However, in the rolling simulation device shown in Patent Document 1, the change is determined to be due to aging deterioration of the rolling mill, and the model formula is updated. As a result, there is a risk that the operation of the rolling line will continue without interruption despite the occurrence of an abnormality.

Furthermore, an inappropriate setting of the control signal input to the rolling mill can also cause the actual measured value of the rolling state to be inappropriate. Even in this case, the rolling simulation device shown in Patent Document 1 updates the model formula without detecting an abnormality in the rolling state of the rolling line.

One objective of the present disclosure is to provide technology that can detect abnormalities in the rolling condition of a rolling line while improving the accuracy of a model that simulates a rolling line.

The first aspect of the present disclosure relates to a rolling emulation device that simulates a rolling line in which multiple rolling machines are lined up in order from the upstream side. The rolling emulation device includes a processor connected to a controller that controls the multiple rolling machines via a communication interface, and a storage device that stores a rolling conveyance model that simulates the conveyance of the rolled material in the rolling line. The processor executes the following processes: acquiring a control signal used to control the multiple rolling machines from the controller; inputting the control signal into the rolling conveyance model to emulate the conveyance of the rolled material; generating a simulated sensor signal that simulates the sensor signal output from multiple sensors provided in the rolling line based on the emulation result by the rolling conveyance model; acquiring actual sensor signals output from the multiple sensors; learning a model correction value that reduces the difference between the simulated sensor signal and the actual sensor signal in the rolling conveyance model; and, if the difference is equal to or greater than a first threshold value, determining that an abnormality has occurred in one of the multiple rolling machines.

The second aspect of the present disclosure relates to a rolling control system that controls a rolling line in which multiple rolling mills are lined up in sequence from the upstream side. The rolling control system includes a controller that controls the multiple rolling mills via a communication interface, a higher-level device that is connected upstream of the controller and generates setting values for control signals to be input to the multiple rolling mills via the controller, and a rolling emulation device that is provided in parallel between the multiple rolling mills and the controller and that emulates the transportation of the rolled material using a rolling transport model that simulates the transportation of the rolled material in the rolling line. When the multiple rolling mills and the controller are not connected to each other and the multiple sensors provided in the rolling line are not connected to each other, the rolling emulation device executes the following processes: acquiring control signals used to control the multiple rolling mills from the controller; inputting the control signals into the rolling transport model to emulate the transportation of the rolled material; generating simulated sensor signals that simulate sensor signals output from the multiple sensors based on the emulation results using the rolling transport model; and outputting the simulated sensor signals to the controller. In addition, the higher-level device executes a determination process to determine whether the rolling state of the rolling line is within an appropriate range based on the simulated sensor signal.

The operation of the above processor may be realized by the processor executing a rolling emulation program.

According to a first aspect of the present disclosure, a control signal used to control a plurality of rolling mills arranged in sequence from the upstream side in a rolling line is acquired from a controller that controls the plurality of rolling mills. The control signal is input to a rolling conveyance model that simulates the conveyance of the rolled material in the rolling line, and the conveyance of the rolled material is emulated. Then, a simulated sensor signal that simulates the sensor signal output from a plurality of sensors provided in the rolling line is generated based on the emulation result by the rolling conveyance model. In addition, in parallel with the generation of the simulated sensor signal, actual sensor signals output from the plurality of sensors are acquired. Then, the rolling conveyance model is learned by reflecting a model correction value for reducing the difference between the simulated sensor signal and the actual sensor signal in the rolling conveyance model. However, if the difference is equal to or greater than a first threshold value, it is determined that an abnormality has occurred in one of the plurality of rolling mills. As a result, it is possible to detect an abnormality in the rolling state of the rolling line while further improving the accuracy of the rolling conveyance model by learning based on the difference between the simulated sensor signal and the actual sensor signal.

According to the second aspect of the present disclosure, when the multiple rolling machines and the controller are not connected to each other and the multiple sensors installed in the rolling line are not connected to each other, a control signal used to control the multiple rolling machines is acquired from the controller. The control signal is input to a rolling conveyance model that simulates the conveyance of the rolled material in the rolling line, and the conveyance of the rolled material is emulated. Then, a simulated sensor signal that simulates the sensor signals output from the multiple sensors is generated based on the emulation result by the rolling conveyance model. Then, based on the simulated sensor signal, it is determined whether the rolling state of the rolling line is within an appropriate range. As a result, when the rolling state of the rolling line 2 becomes inappropriate, it is possible to detect an abnormality in the rolling state of the rolling line and to check whether the set value of the control signal is appropriate. Furthermore, when a part of the rolling line is changed or a new rolling line is installed, the operation of the rolling line is started after confirming that the set value of the control signal is appropriate. Therefore, the frequency with which the set values of the control signals input to a plurality of rolling mills need to be adjusted during operation of the rolling line is reduced, and this is expected to improve the finishing accuracy of the rolled material.

FIG. 1 is a diagram for explaining an overview of a rolling control system according to a first embodiment. 1 is a block diagram showing a configuration example of a rolling emulation device according to a first embodiment; 2 is a block diagram showing a processing example of the rolling emulation device according to the first embodiment; FIG. 4 is a flowchart showing an example of processing performed by the rolling emulation device according to the first embodiment; FIG. 11 is a diagram for explaining an overview of a rolling control system according to a second embodiment. FIG. 11 is a block diagram showing a processing example of a rolling control system according to a second embodiment. 10 is a flowchart showing an example of processing of a rolling control system according to a second embodiment.

The rolling control system, rolling emulation device, and rolling emulation program according to the embodiments of the present disclosure will be described with reference to the attached drawings. In addition, elements common to each drawing will be assigned the same reference numerals and duplicated descriptions will be omitted.

1. First embodiment
1-1. Overview of the rolling control system Fig. 1 is a diagram for explaining an overview of a rolling control system 1A according to the first embodiment. The rolling control system 1A controls a rolling line 2. The rolling line 2 is, for example, a hot rolling line or a cold rolling line. The rolling line 2 includes a plurality of rolling mills 4 arranged in parallel along the transport direction of the material 3 to be rolled. In other words, the rolling line 2 has a plurality of rolling mills 4 arranged in sequence from the upstream side.

The rolling line 2 further includes a plurality of sensors 6. The plurality of sensors 6 measure the rolling state of the rolled material 3. The plurality of sensors 6 may be provided, for example, on the input side of the first (most upstream) rolling mill 4 in the conveying direction of the rolled material 3, on the output side of the last (most downstream) rolling mill 4 in the conveying direction of the rolled material 3, or between the rolling mills 4. In other words, the plurality of sensors 6 may be disposed in various locations. Examples of the plurality of sensors 6 include a plate speed meter, a plate thickness meter, etc. The information acquired by the plurality of sensors 6 is input to the controller 8. The information acquired by the sensors 6 is also referred to as an actual sensor signal.

Each of the multiple rolling mills 4 includes an actuator 7 and is driven by the actuator 7. An actuator 7 is provided for each rolling mill 4. The actuator 7 controls its drive in accordance with a control signal output from a controller 8. The control signal output from the controller 8 is, for example, a parameter input to the actuator 7. Examples of parameters include the rotation speed of the rolls included in the rolling mill 4 and the opening degree of the rolls. The parameters input to the actuator 7 may be generated by the controller 8 or may be generated by the higher-level device 50.

The rolling recording system 1 comprises the above-mentioned rolling line 2, a controller 8, a rolling emulation device 10, and a higher-level device 50.

The controller 8 controls the multiple rolling mills 4 via a communication interface. Specifically, the controller 8 executes sequence control of the control signals input to the actuators 7 and control to transmit predetermined information to the higher-level device 50. The predetermined information includes actual sensor signals acquired by the sensors 6 (such as actual measured values of the rolling state). Furthermore, the predetermined information may include information related to the control signals (such as information on the control signals input to the actuators 7).

The controller 8 may, for example, include one or more processors (not shown) and one or more storage devices (not shown), or may include one or more dedicated hardware (not shown). As a specific example, the controller 8 is a PLC (Programmable Logic Controller).

The higher-level device 50 manages the rolling state of the rolling line 2. Specifically, the higher-level device 50 generates a setting value (parameter) of the control signal to be input to the rolling mill 4 (actuator 7). The setting value of the control signal may be generated based on a simulation result analyzed by a dedicated tool for simulating the rolling state of the rolling line 2, or may be generated based on a difference value between a predicted value of the rolling state based on the simulation result and an actual sensor signal. The setting value of the control signal is also simply referred to as the control signal.

The upper level device 50 is connected upstream of the controller 8. The upper level device 50 further includes, for example, one or more processors (not shown) and one or more storage devices (not shown).

The rolling emulation device 10 emulates the transportation of the rolled material 3 in the rolling line 2. A rolling transport model (not shown) that simulates the transportation of the rolled material 3 is used to emulate the transportation of the rolled material 3. The rolling transport model here is premised on a model that has undergone a certain amount of learning. Based on this, for example, if the rolling state obtained from the rolling transport model differs from the actual rolling state, the rolling emulation device 10 judges whether it is necessary to further improve the accuracy of the rolling transport model, whether there is an abnormality in the rolling machine 4, etc. This makes it possible to detect abnormalities in the rolling state of the rolling line 2 while further improving the accuracy of the rolling transport model.

The outline of the processing by the rolling emulation device 10 is as follows. The rolling emulation device 10 executes the following: a process of acquiring a control signal used to control the multiple rolling mills 4 from a controller 8 that controls the multiple rolling mills 4 arranged in sequence from the upstream side in the rolling line 2; a process of inputting a control signal to a rolling conveyance model to emulate the conveyance of the rolled material 3; a process of generating a simulated sensor signal that simulates the sensor signal output from the multiple sensors 6 provided in the rolling line 2 based on the emulation result by the rolling conveyance model; a process of acquiring the actual sensor signal output from the multiple sensors 6; a learning process of reflecting a model correction value for reducing the difference between the simulated sensor signal and the actual sensor signal in the rolling conveyance model; and an abnormality determination process of determining that an abnormality has occurred in one of the multiple rolling mills if the difference is equal to or greater than a first threshold value. Details of the processing by the rolling emulation device 10 will be described later.

Consider the location where the rolling emulation device 10 is installed. In the rolling emulation device 10, a process is performed to compare the difference between the simulated sensor signal and the actual sensor signal. In this case, it is desirable that the difference between the time when the simulated sensor signal is generated in the rolling emulation device 10 and the time when the rolling emulation device 10 acquires the actual sensor signal is small. On the other hand, it is assumed that the time from when the set value of the control signal is input to the rolling mill 4 to when the actual sensor signal is output by the sensor 6 is short. For this reason, it is desirable that the rolling emulation device 10 is installed downstream of the controller 8, similar to the rolling mill 4. Therefore, the rolling emulation device 10 is installed in parallel between the multiple rolling mills 4 and the controller 8.

1-2. Example of a Rolling Emulation Apparatus 1-2-1. Configuration Example FIG. 2 is a block diagram showing a configuration example of the rolling emulation apparatus 10 according to the first embodiment.

The rolling emulation device 10 includes one or more processors 20 (hereinafter simply referred to as processor 20) and one or more storage devices 30 (hereinafter simply referred to as storage devices 30). The processor 20 executes various processes. An example of the processor 20 is a CPU (Central Processing Unit). The storage device 30 stores various information required for processing by the processor 20. Examples of the storage device 30 include a volatile memory, a non-volatile memory, a HDD (Hard Disk Drive), an SSD (Solid State Drive), etc.

The various information stored in the storage device 30 includes a rolling conveying model 31 and a rolling emulation program. As described above, the rolling conveying model 31 is a model that simulates the conveyance of the rolled material 3 on the rolling line 2. The rolling emulation program (not shown) is a computer program executed by the processor 20. The various functions of the rolling emulation device 10 may be realized by the processor 20 executing the rolling emulation program. The rolling emulation program is not limited to being stored in the storage device 30, and may be recorded in a computer-readable storage medium.

1-2-2. Processing Example Fig. 3 is a block diagram showing a processing example of the rolling emulation device 10 according to the embodiment.

Specifically, the rolling emulation device 10 (more specifically, the processor 20) inputs a control signal output from the controller 8. Then, the rolling emulation device 10 executes an emulation process 11 that emulates the transportation of the rolled material 3 based on the control signal. In the emulation process 11, a process is performed to read a rolling transportation model 31 stored in the storage device 30. Furthermore, in the emulation process 11, the rolling emulation device 10 generates simulated sensor signals that simulate the sensor signals output from the multiple sensors 6 based on the emulation results by the rolling transportation model 31.

Furthermore, the rolling emulation device 10 acquires actual sensor signals output from the multiple sensors 6. The rolling emulation device 10 then performs a difference determination process 12 based on the simulated sensor signals and the actual sensor signals. In the difference determination process 12, the rolling emulation device 10 determines whether the difference between the simulated sensor signal and the actual sensor signal is less than a first threshold value. If the difference is less than the first threshold value, the rolling emulation device 10 performs a model correction value generation process 13 to generate a model correction value to reduce the difference. After performing the model correction value generation process 13, the rolling emulation device 10 performs a learning process to reflect the model correction value in the rolling conveyance model 31. This allows the accuracy of the rolling conveyance model 31 to be further improved.

When the difference is small, it is considered that the rolling conveying model has a certain level of accuracy. In this case, the rolling emulation device 10 may not reflect the difference in the rolling conveying model. Therefore, in the learning process, the rolling emulation device 10 may reflect the model correction value in the rolling conveying model 31 when the difference is less than the first threshold value and is equal to or greater than a second threshold value that is smaller than the first threshold value.

On the other hand, if the difference is equal to or greater than the first threshold value, the rolling emulation device 10 executes an abnormality determination process in the difference determination process 12 to determine that an abnormality has occurred in one of the multiple rolling mills 4. In this case, the rolling emulation device 10 notifies the higher-level device 50 via the controller 8 that an abnormality has occurred in the rolling mill 4. This makes it possible to detect an abnormality in the rolling state of the rolling line 2. An abnormality in the rolling mill 4 means, for example, that the amount of change in a parameter indicating the state of the rolling mill 4 is equal to or greater than a predetermined value. Examples of factors that cause the amount of change in the parameter to be equal to or greater than a predetermined value include aging deterioration of the rolling mill 4, a breakdown of the rolling mill 4, etc.

Furthermore, after executing the difference determination process 12, the rolling emulation device 10 may perform a storage process 14 to store the difference value between the simulated sensor signal and the actual sensor signal and the state of various sensor signals. This makes it possible to check the history of the difference value and the state of various sensor signals when a cause for the rolling state of the rolling line 2 becoming inappropriate occurs. This makes it easier to identify the cause for the rolling state of the rolling line becoming inappropriate.

Figure 4 is a flowchart summarizing an example of processing by the rolling emulation device 10.

In step S100, the rolling emulator 10 acquires various information. Then, the process proceeds to step S110. The various information includes a control signal output from the controller 8 and actual sensor signals output from the multiple sensors 6.

In step S110, the rolling emulation device 10 emulates the transportation of the rolled material 3 using the rolling transportation model 31. Then, the process proceeds to step S120.

In step S120, the rolling emulation device 10 generates simulated sensor signals that simulate the sensor signals output from the multiple sensors 6 based on the emulation results by the rolling conveyance model 31. Then, the process proceeds to step S130.

In step S130, the rolling emulator 10 calculates the difference between the simulated sensor signal and the actual sensor signal. Then, the process proceeds to step S140.

In step S140, the rolling emulation device 10 determines whether the difference between the simulated sensor signal and the actual sensor signal is less than a threshold value. If it is determined that the difference is less than the threshold value (step S140; Yes), the process proceeds to step S150. Otherwise (step S140; No), the process proceeds to step S170.

In step S150, the rolling emulation device 10 generates a model correction value to reduce the difference. Then, the process proceeds to step S160.

In step S160, the rolling emulation device 10 performs a learning process to reflect the model correction value in the rolling conveying model 31.

Furthermore, as described above, if the difference is less than the first threshold value and is greater than or equal to a second threshold value that is smaller than the first threshold value, the rolling emulation device 10 may reflect the model correction value in the rolling conveying model 31.

In step S170, the rolling emulation device 10 determines that an abnormality has occurred in one of the multiple rolling mills 4. Processing then proceeds to step S180.

In step S180, the rolling emulation device 10 notifies the higher-level device 50 via the controller 8 that an abnormality has occurred in the rolling mill 4.

1-3. Effects According to this embodiment, a control signal used to control the rolling mills 4 is acquired from a controller 8 that controls the rolling mills 4 arranged in sequence from the upstream side in the rolling line 2. The control signal is input to a rolling conveyance model 31 that simulates the conveyance of the material 3 to be rolled in the rolling line 2, and the conveyance of the material 3 to be rolled is emulated. Then, based on the emulation result by the rolling conveyance model 31, a simulated sensor signal that simulates the sensor signal output from the multiple sensors 6 provided in the rolling line 2 is generated. In parallel with the generation of the simulated sensor signal, the actual sensor signal output from the multiple sensors 6 is acquired. Then, a model correction value for reducing the difference between the simulated sensor signal and the actual sensor signal is reflected in the rolling conveyance model 31, and the rolling conveyance model 31 is trained. However, if the difference is equal to or greater than the first threshold value, it is determined that an abnormality has occurred in any one of the multiple rolling mills 4. This makes it possible to detect abnormalities in the rolling state of the rolling line 2 while further improving the accuracy of the rolling conveyance model 31 through learning based on the difference between the simulated sensor signal and the actual sensor signal.

2. Second embodiment
2-1. Overview of the rolling control system FIG. 5 is a diagram for explaining the overview of the rolling control system 1B according to the second embodiment. For example, consider a case where the cause of the rolling state of the rolling line 2 being inappropriate is not the accuracy of the rolling conveyance model or an abnormality in the rolling mill 4, but the set value of the control signal input to the rolling mill 4. In this case, the rolling emulation device 10 of the rolling control system 1A according to the first embodiment described above cannot identify the cause. Therefore, in the rolling control system 1B, when the rolling state of the rolling line 2 is inappropriate, it is determined whether the set value of the control signal input to the multiple rolling mills 4 is appropriate.

The basic configuration of the rolling control system 1B is the same as that of the rolling control system 1A described above. The differences from the rolling control system 1A include, as shown in FIG. 5, that the multiple rolling mills 4 and the controller 8 are in a disconnected state (first disconnected state), that the multiple sensors 6 and the controller 8 are in a disconnected state (second disconnected state), and that the simulated sensor signal generated by the rolling emulation device 10 is output to the controller 8.

The first non-connected state is, for example, a state in which the wiring between the multiple rolling mills 4 and the controller 8 is cut off, a state in which the use of control signals input to the rolling mills 4 (actuators 7) is restricted, etc. The second non-connected state is, for example, a state in which the wiring between the multiple sensors 6 and the controller 8 is cut off, a state in which the output of sensor signals from the sensors 6 is restricted, etc.

Switching between the rolling control system 1A and the rolling control system 1B is controlled by the higher-level device 50. For example, when the operation mode of the rolling line 2 is the operation mode, the rolling control system 1A is selected, and when the operation mode of the rolling line 2 is the inspection mode, the rolling control system 1B is selected. Furthermore, when the rolling control system 1B is selected, switching to the first non-connected state and switching to the second non-connected state are performed.

The outline of the processing by the rolling control system 1B is as follows. When the multiple rolling mills 4 and the controller 8 are not connected and the multiple sensors 6 and the controller 8 are not connected, the rolling control system 1B executes the following processes: acquiring control signals used to control the multiple rolling mills 4 from the controller 8 that controls the multiple rolling mills 4; inputting a control signal to the rolling conveyance model 31 to emulate the conveyance of the material to be rolled 3; generating simulated sensor signals that simulate the sensor signals output from the multiple sensors 6 installed in the rolling line 2 based on the emulation results by the rolling conveyance model 31; outputting the simulated sensor signals to the controller 8; and determining whether the rolling state of the rolling line 2 is within an appropriate range based on the simulated sensor signals. Details of the processing by the rolling control system 1B will be described later.

6 is a block diagram showing a processing example of the rolling control system 1B according to the embodiment 2. The processing of the rolling control system 1B includes the processing of the rolling emulator device 10 and the processing of the higher-level device 50.

An example of processing of the rolling emulation device 10 will be described. Specifically, the rolling emulation device 10 (processor 20) inputs a control signal output from the controller 8. Then, the rolling emulation device 10 executes an emulation process 11 that emulates the transportation of the rolled material 3 based on the control signal. Furthermore, in the emulation process 11, the rolling emulation device 10 generates a simulated sensor signal that simulates the sensor signals output from the multiple sensors 6 based on the emulation result by the rolling transportation model 31. Then, the rolling emulation device 10 outputs the simulated sensor signal to the controller 8.

Next, an example of the processing of the upper device 50 will be described. Specifically, the upper device 50 executes a process for generating a set value of the control signal (process for generating a set value of the control signal 51). After executing the process for generating a set value of the control signal 51, the upper device 50 outputs the control signal to the rolling emulator device 10 via the controller 8. Furthermore, the upper device 50 acquires a simulated sensor signal output from the rolling emulator device 10 via the controller 8. Then, the upper device 50 executes a process for determining whether the rolling state of the rolling line 2 is within an appropriate range (process for determining the rolling state 52) based on the simulated sensor signal. After that, if it is determined that the rolling state of the rolling line 2 is not within the appropriate range, the upper device 50 executes a process for changing the set value of the control signal (process for changing the set value of the control signal 53). The rolling state determination process 52 includes an abnormality determination process as shown in FIG. 6. The abnormality determination process means a process for determining that an abnormality has occurred in any one of the rolling mills 4 among the multiple rolling mills 4 when it is determined that the rolling state of the rolling line 2 is not within the appropriate range. This makes it possible to detect abnormalities in the rolling state of the rolling line 2.

Figure 7 is a flowchart summarizing an example of processing by the rolling control system 1B.

In step S200, the rolling control system 1B generates a set value for the control signal. Then, the process proceeds to step S210.

In step S210, the rolling control system 1B inputs a control signal to the rolling conveying model 31 to emulate the conveying of the rolled material 3. Then, the process proceeds to step S220.

In step S220, the rolling control system 1B generates simulated sensor signals that simulate the sensor signals output from the multiple sensors 6 based on the emulation results by the rolling conveyance model 31. Then, the process proceeds to step S230.

In step S230, the rolling control system 1B determines whether the rolling state of the rolling line 2 is within the appropriate range. If it is determined that the rolling state of the rolling line 2 is within the appropriate range (step S230; Yes), the processing ends. Otherwise (step S230; No), the processing proceeds to step S240.

In step S240, the rolling control system 1B changes the setting value of the control signal.

2-3. Effects According to this embodiment, when the plurality of rolling mills 4 and the controller 8 are not connected to each other and the plurality of sensors 6 provided in the rolling line 2 are not connected to each other, a control signal used to control the plurality of rolling mills 4 is acquired from the controller 8. The control signal is input to a rolling conveyance model 31 that simulates the conveyance of the material 3 to be rolled in the rolling line 2, and the conveyance of the material 3 to be rolled is emulated. Then, based on the emulation result by the rolling conveyance model 31, a simulated sensor signal that simulates the sensor signal output from the plurality of sensors 6 is generated. Then, based on the simulated sensor signal, it is determined whether the rolling state of the rolling line 2 is within an appropriate range. As a result, when the rolling state of the rolling line 2 becomes inappropriate, it is possible to detect an abnormality in the rolling state of the rolling line 2 and to check whether the set value of the control signal is appropriate.

Furthermore, when changing part of the rolling line 2 or installing a new rolling line 2, the operation of the rolling line 2 is started after confirming that the setting values of the control signals are appropriate. Therefore, while the rolling line 2 is in operation, the frequency of adjusting the setting values of the control signals input to the multiple rolling mills 4 is reduced, and the finishing accuracy of the rolled material 3 can be expected to improve.

1A, 1B...Rolling control system, 2...Rolling line, 3...Rolled material, 4...Rolling mill, 6...Sensor, 7...Actuator, 8...Controller, 10...Rolling emulation device, 11...Emulation processing, 12...Difference determination processing, 13...Model correction value generation processing, 14...Storage processing, 20...Processor, 30...Memory device, 31...Rolling transport model, 32...Rolling emulation program, 50...Higher-level device, 51...Generation processing of control signal setting values, 52...Rolling state determination processing, 53...Change processing of control signal setting values

Claims (6)

A rolling emulation device that simulates a rolling line in which a plurality of rolling mills are arranged in sequence from the upstream side,
a processor connected to a controller that controls the plurality of rolling mills via a communication interface;
A storage device storing a rolling transport model simulating the transport of the rolled material in the rolling line,
The processor,
A process of acquiring a control signal used to control the plurality of rolling mills from the controller;
A process of inputting the control signal into the rolling conveyance model to emulate the conveyance of the rolled material;
A process of generating simulated sensor signals that simulate sensor signals output from a plurality of sensors provided in the rolling line based on an emulation result using the rolling conveyance model;
acquiring actual sensor signals output from the plurality of sensors;
A learning process for reflecting a model correction value for reducing a difference between the simulated sensor signal and the actual sensor signal in the rolling conveyance model;
an abnormality determination process for determining that an abnormality has occurred in any one of the rolling mills when the difference is equal to or greater than a first threshold value;
23. A rolling emulation apparatus configured to execute the above steps. 2. The rolling emulation apparatus according to claim 1,
The processor further comprises:
a process of notifying a host device connected upstream of the controller that an abnormality has occurred in the rolling mill when the difference is equal to or greater than the first threshold value;
23. A rolling emulation apparatus configured to execute the above steps. 3. The rolling emulation device according to claim 1,
The processor, in the learning process,
A rolling emulation device characterized in that, when the difference is less than a first threshold value and is equal to or greater than a second threshold value smaller than the first threshold value, the model correction value is reflected in the rolling transportation model. A rolling emulation program simulating a rolling line in which a plurality of rolling mills are arranged in sequence from the upstream side,
A process of acquiring control signals used to control the plurality of rolling mills from a controller that controls the plurality of rolling mills via a communication interface;
A process of inputting the control signal into a rolling conveyance model that simulates the conveyance of the rolled material in the rolling line to emulate the conveyance of the rolled material;
A process of generating simulated sensor signals that simulate sensor signals output from a plurality of sensors provided in the rolling line based on an emulation result using the rolling conveyance model;
acquiring actual sensor signals output from the plurality of sensors;
A learning process for reflecting a model correction value for reducing a difference between the simulated sensor signal and the actual sensor signal in the rolling conveyance model;
an abnormality determination process for determining that an abnormality has occurred in any one of the rolling mills when the difference is equal to or greater than a first threshold value;
configured to cause a computer to execute
A rolling emulation program comprising: A rolling control system for controlling a rolling line in which a plurality of rolling mills are arranged in sequence from an upstream side, comprising:
A controller that controls the plurality of rolling mills via a communication interface;
a host device connected upstream of the controller and generating setting values of control signals to be input to the plurality of rolling mills via the controller;
a rolling emulation device that is provided in parallel between the plurality of rolling mills and the controller, and that emulates the transportation of the rolled material using a rolling transportation model that simulates the transportation of the rolled material in the rolling line;
Equipped with
When the plurality of rolling mills and the controller are not connected to each other and when the plurality of sensors provided in the rolling line and the controller are not connected to each other, the rolling emulator device
A process of acquiring a control signal used to control the plurality of rolling mills from the controller;
A process of inputting the control signal into the rolling conveyance model to emulate the conveyance of the rolled material;
A process of generating simulated sensor signals that simulate sensor signals output from the plurality of sensors based on an emulation result using the rolling conveyance model;
outputting the simulated sensor signal to the controller;
configured to run
The rolling control system according to the present invention is characterized in that the host device is configured to execute a determination process for determining whether or not the rolling state of the rolling line is within an appropriate range based on the simulated sensor signal. The rolling control system according to claim 5,
The rolling control system is characterized in that the host device is configured to change the set value of the control signal when the rolling state is not within the appropriate range in the judgment process.

Images