Process management system and process management method

Abstract

To predict dynamic optimum control for a process to be controlled and to monitor the state change of the process due to the predicted optimum control in plant operation support.SOLUTION: The process management system generates control prediction data related to a control flow for a process to be controlled, based on a control prediction model that outputs, in response to an input of a target value of a physical quantity in the process, the control flow including operations and an order of the operations for the process until the target value is reached. Then, the process management system generates process prediction data related to a future predicted value of the physical quantity in a case in which the process to be controlled is controlled on the basis of the control prediction data on the basis of a process prediction model that outputs the future predicted value of the physical quantity in the process in response to an input of the control prediction data and the control prediction data. Then, the process management system outputs information related to the control prediction data and the process prediction data from an output device in association with each other.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present invention relates to a process management system and a process management method. [Background technology]

[0002] Conventionally, plants such as petrochemical plants, power plants, and water treatment plants have been controlled by a DCS (Distributed Control System), which connects the equipment that makes up the plant with the control devices that control the equipment. Furthermore, the DCS collects and records performance data related to the automatic control of the plant and the manual control of the plant through the intervention of workers.

[0003] For use in controlling a plant, a model may be created of the relationship between operations performed on the plant and changes in the plant's state based on past performance data, etc. Related prior art includes the technology disclosed in Patent Document 1.

[0004] Furthermore, in order to predict the future state of the plant, a model for predicting future physical quantities in the plant may be constructed. Related prior art includes the technology disclosed in Patent Document 2. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2022-3493 [Patent Document 2] Japanese Patent Application Publication No. 2023-80580 Summary of the Invention [Problem to be solved by the invention]

[0006] For example, to identify the optimal control for a process, it is possible to use a model of the relationship between the operations performed on the plant and changes in the plant's state. However, the process in the plant may change due to various factors such as adjustments to operating conditions and environmental changes, so it is desirable to be able to predict and monitor how the process will change under the identified control.

[0007] For example, a plant may be predicted and monitored using a model for predicting future physical quantities of a controlled process. However, changes in the plant process can be complex, and it may be difficult to efficiently or accurately identify control measures for achieving a target state using such a model alone.

[0008] For these reasons, in order to improve plant operation, a system is needed that can efficiently identify optimal control for the target state of the process to be controlled, while at the same time predicting and monitoring changes in the state of the process due to optimal control.

[0009] The above-mentioned problems are not taken into consideration in Patent Documents 1 and 2.

[0010] The present invention has been made in consideration of the above points, and aims to predict dynamic optimal control for a process to be controlled in plant operation support, and to monitor changes in the state of the process due to the predicted optimal control. [Means for solving the problem]

[0011] In order to achieve the above-mentioned object, one aspect of the present invention is a process management system that manages a process executed in equipment constituting a plant, the process management system having a processor and a memory, the processor generates control prediction data related to a control flow for the process to be controlled, based on a control prediction model that, in response to an input of a target value of a physical quantity in the process, outputs a control flow including operations for the process until the target value is reached and an order of the operations, the processor generates process prediction data related to a future predicted value of the physical quantity when the process to be controlled is controlled based on the control prediction data, based on the process prediction model and the control prediction data, and information related to the control prediction data is output from an output device in association with the control prediction data. [Effects of the Invention]

[0012] According to the present invention, for example, in plant operation support, it is possible to predict dynamic optimal control for a process to be controlled, and to monitor changes in the state of the process due to the predicted optimal control. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a diagram showing the configuration of a process management system according to an embodiment. [Figure 2] FIG. 4 is a diagram showing driving performance data according to the embodiment. [Figure 3] FIG. 4 is a diagram showing characteristic information according to the embodiment. [Figure 4] FIG. 2 is a diagram showing a state transition model according to the embodiment. [Figure 5] FIG. 4 is a diagram showing a control prediction model according to the embodiment. [Figure 6] FIG. 4 is a diagram showing control prediction data according to the embodiment. [Figure 7] FIG. 4 is a diagram showing a process prediction model and characteristic parameters according to the embodiment. [Figure 8] FIG. 10 is a diagram showing a prediction process according to the embodiment. [Figure 9] FIG. 1 is a diagram showing use case 1 (real-time processing) of the prediction processing according to the embodiment. [Figure 10] FIG. 10 is a diagram showing a GUI (real-time processing) according to Use Case 1 of the embodiment. [Figure 11] FIG. 10 is a diagram showing a GUI (real-time processing: comparison with the initial setting flow) according to Use Case 1 of the embodiment. [Figure 12] FIG. 10 is a diagram showing use case 2 (simulation processing) of the embodiment. [Figure 13] FIG. 10 is a diagram showing a GUI (simulation process) according to a second use case of the embodiment. [Figure 14] FIG. 10 is a diagram showing use case 3 (automatic control) of the embodiment. [Figure 15] FIG. 4 is a diagram showing a control monitoring process according to the embodiment. [Figure 16] FIG. 10 is a diagram illustrating a use case 4 (model update) according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments are examples for explaining the present invention, and appropriate omissions and simplifications have been made for clarity of explanation. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. The position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc., in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

[0015] Examples of various types of information may be described using expressions such as "table," "list," and "queue," but the various types of information may also be expressed using data structures other than these. For example, various types of information such as "XX table," "XX list," and "XX queue" may also be expressed as "XX information." When describing identification information, expressions such as "identification information," "identifier," "name," "ID," and "number" are used, but these are interchangeable.

[0016] When there are multiple components with the same or similar functions, they may be described using the same reference numeral with different subscripts. When there is no need to distinguish between these multiple components, the subscripts may be omitted.

[0017] In the embodiments, processing performed by executing a program may be described. Here, a computer executes the program using a processor (e.g., a CPU or a GPU) and performs processing defined by the program using storage resources (e.g., a memory) and interface devices (e.g., a communication port). Therefore, the entity performing the processing by executing the program may be the processor. Similarly, the entity performing the processing by executing the program may be a controller, device, system, computer, or node having a processor. The entity performing the processing by executing the program may be any computing unit, and may include a dedicated circuit that performs specific processing. Here, the dedicated circuit may be, for example, an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), or a CPLD (Complex Programmable Logic Device).

[0018] A program may be installed on a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server may include a processor and storage resources for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. In addition, in an embodiment, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0019] [Embodiment] FIG. 1 is a diagram showing the configuration of a process control system 1 according to an embodiment. The plant whose process is to be controlled by the process control system 1 is assumed to be a non-steady-state plant in which target values ​​of physical quantities measured by equipment constituting the plant change over time. However, the plant is not limited to a non-steady-state plant, and the plant may also be one in which target values ​​of physical quantities do not change over time. Furthermore, although the process to be controlled is assumed to be a batch process, it may also be a continuous process.

[0020] The process control system 1 includes a memory 11, an input device 12, an output device 13, a processor 14, a storage unit 15, and a communication interface 16. The process control system 1 is connected via a control device 2 to one or more pieces of equipment 3 for each process that constitutes a plant.

[0021] The control device 2 acquires operation record data 15a of the equipment 3, which includes information on measured values ​​of physical quantities of the process executed in the equipment 3, and transmits the data to the process management system 1. The control device 2 also transmits control instructions for the equipment 3 received from the process management system 1 to the equipment 3.

[0022] The memory 11 of the process control system 1 is a storage device having a volatile storage area. The input device 12 is a device such as a keyboard or mouse for inputting operations by the administrator to the process control system 1. The output device 13 is a device such as a display or speaker for outputting information from the process control system 1 to the administrator. The processor 14 executes programs in cooperation with the memory 11. The storage unit 15 is a storage unit having a non-volatile storage area. The communication interface 16 is a communication device that enables the process control system 1 to communicate with the control device 2.

[0023] The processor 14 includes a data collection unit 14a, a control instruction unit 14b, a control prediction unit 14c, a process prediction unit 14d, a model evaluation unit 14e, and a model creation / update unit 14f, which are realized by executing a program.

[0024] The data collection unit 14a collects operation record data 15a, converts it into an appropriate format, and stores it in the storage unit 15. The control instruction unit 14b outputs a command to control each facility device 3 to the control device 2 based on the current control prediction data 15d.

[0025] The control prediction unit 14c performs control prediction based on the operation record data 15a or on hypothetical conditions using the control prediction model 15c. The process prediction unit 14d performs control prediction based on the operation record data 15a or on hypothetical conditions using the process prediction model 15e.

[0026] The model evaluation unit 14e calculates evaluation information 15h on the model accuracy of the control prediction model 15c and the process prediction model 15e by comparing the prediction results of the control prediction model 15c and the process prediction model 15e with the operation record data 15a.

[0027] The model creation and updating unit 14f selects learning data from the operation record data 15a, learns the selected learning data, and creates a control prediction model 15c and a process prediction model 15e. The model creation and updating unit 14f also selects learning data from the operation record data 15a so as to maximize accuracy under any conditions, and learns the selected learning data to update the control prediction model 15c and the process prediction model 15e.

[0028] The storage unit 15 stores operation record data 15a, characteristic information 15b, a control prediction model 15c, control prediction data 15d, a process prediction model 15e, process prediction data 15f, characteristic parameters 15g, and evaluation information 15h.

[0029] The control prediction model 15c and the process prediction model 15e are provided for each characteristic. The evaluation information 15h is information about the accuracy of the control prediction model 15c and the process prediction model 15e.

[0030] The control prediction model 15c is a model that outputs the operation amount of the facility device 3 to reach the target value based on information on the target value of the physical quantity in the plant process and information on the measurement value of the physical quantity related to the process. The control prediction model 15c is a model that has learned the relationship between information on the operation amount of the facility device 3 included in past operation records and information on the measurement value of the operation result. The control prediction model 15c has learned multiple state transition models 15c0 until the measurement value of the operation result reaches the target value of the physical quantity.

[0031] The control prediction model 15c can select an optimal operation flow (control flow) from among a plurality of state transition models 15c0 until the measured value of the operation result reaches the target value of the physical quantity. In any case, if any constraint conditions (reaching the target value faster, the measured value not exceeding a certain value, reducing the amount of operation, etc.) are satisfied, the corresponding operation flow is defined as the "optimal operation flow."

[0032] The process prediction model 15e includes two use cases. Use case 1 is a case in which the future is predicted based on the current state. Use case 2 is a case in which the future is predicted based on a virtual state (simulation mode or learning data expansion mode of the control prediction model 15c).

[0033] In use case 1, the input values ​​of the process prediction model 15e are the measured values ​​of the physical quantities (PV values, present values) plus the operation flow, and the output values ​​are the predicted values ​​of the physical quantities in the future. Specifically, in use case 1, the future predicted values ​​of the physical quantities related to the process are output based on information on the measured values ​​of the physical quantities related to the process and information on the operation quantities of the facility devices 3.

[0034] The process prediction model 15e includes a physical model. The physical model makes it possible to output a predicted value even when there is little learning data to refer to. The process prediction model 15e also includes a machine learning model. The machine learning model makes it possible to output a predicted value based on data even in cases where a physical principle equation cannot be established or where disturbances occur.

[0035] In use case 2, the input values ​​of the process prediction model 15e are virtual data (PV values) of physical quantities plus an operation flow, and the output values ​​are predicted values ​​of future physical quantities.

[0036] (Driving performance data 15a according to the embodiment) 2 is a diagram showing operation record data 15a according to the embodiment. The operation record data 15a has columns of "operation No.", "product type", "facility / device", "date and time", "operation", and "measured value".

[0037] "Operation No." is identification information of an operation performed by a plant operator. "Product type" is the product type of the object (e.g., a manufactured product) handled by the equipment 3 to be operated. "Equipment" is identification information of the equipment 3 to be operated. "Date and time" is the date and time when the operation in question was performed. "Operation" indicates the operation content of the operator. "Measured value" is the value of a physical quantity such as flow rate, pressure, or temperature measured by the equipment 3 in question. "Measured value" includes values ​​of one or more types of physical quantities. By classifying the "measured value" into multiple discrete "states," one of the multiple "states" is associated with an "operation" in each record of the operation performance data 15a.

[0038] (Characteristic information 15b according to the embodiment) Fig. 3 is a diagram showing characteristic information 15b according to the embodiment. The characteristic information 15b stores characteristics of the controlled object (for example, product type, facility equipment, etc.) and characteristic types (A, B, A, B, etc.) included in each characteristic in association with each other. Fig. 3 shows that, for example, A, B, etc. are defined as characteristic types for the product type, and A, B, etc. are defined as characteristic types for the facility equipment 3.

[0039] (State transition model 15c0 according to the embodiment) 4 is a diagram illustrating a state transition model 15c0 according to the embodiment. The state transition model 15c0 is a model that serves as a basis for creating the control prediction model 15c. The state transition model 15c0 is created for each type of equipment and each piece of facility equipment 3.

[0040] 4 shows that when operation a1 is performed on facility device 3 that is the operation target and in state s190, the probability of transitioning to state s211 at a predetermined transition time (e.g., after 10 seconds) is 20%, the probability of transitioning to state s212 is 70%, and the probability of transitioning to state s213 is 10%. Also, when operation a2 is performed in the same state s190, the probability of transitioning to state s233 is 55%, and the probability of transitioning to state s234 is 45%. This state transition model 15a1 can be created by discretizing past operation performance data 15a into "states" and then statistically processing the relationship between the operation (setting value of the physical quantity) and the state after 10 seconds, which is the transition time.

[0041] (Control prediction model 15c according to the embodiment) FIG. 5 is a diagram showing a control prediction model 15c according to the embodiment.

[0042] The control prediction model 15c is a model that represents the optimal action (operation) to be taken in order to reach a target state from a certain state as quickly as possible. The control prediction model 15c can be obtained from the state transition model 15c0 using reinforcement learning or dynamic programming. When reinforcement learning is used, a large reward is given when performing a certain operation from a certain state makes it easier to approach the target state.

[0043] The control prediction model 15c stores information about a second state to which a control object in a first state transitions when a first operation is performed on the control object. For example, the example shown in the first row of Fig. 5 indicates that in the case of state s190, performing operation a2 is the optimal operation, resulting in a transition to state s233.

[0044] The state transition model 15c0 in Fig. 4 represents multiple operations that have been performed in the past for a certain state and the transition probability when each operation is performed. In contrast, the control prediction model 15c in Fig. 5 represents only the optimal operation, among multiple operations that have been performed in the past, for ultimately transitioning to the target state.

[0045] As described above, in the learning phase of the control prediction model 15c, the state transition model 15c0 is created from the past driving performance data 15a, and the control prediction model 15c is created from the state transition model 15c0. Note that the control prediction model 15c may store information required for transitioning from a first state to a second state for each combination of the first state and the second state.

[0046] The control prediction model 15c is learned as follows. First, the processor 14 acquires operation record data 15a relating to the operation record of the equipment 3, including measured values ​​of physical quantities in the process. Next, the processor 14 determines the state of the process from the measured values ​​of the physical quantities. Next, the processor 14 classifies the determined process states into multiple categories using a predetermined classification process (clustering, etc.). Next, based on the process states classified into multiple categories, the processor 14 generates a state transition model 15c0 including one or more states of the process that are transition destination states to which the process transitions when a certain operation is performed on a process in a certain state, and the transition probability to this state. Next, the processor 14 generates the control prediction model 15c by generating an optimal transition route for the process state to transition from a first state to a second state based on the state transition model 15c0.

[0047] (Control prediction data 15d according to the embodiment) Fig. 6 is a diagram showing control prediction data 15d according to an embodiment. Fig. 6 shows control prediction data 15d for a transition from state s190, which is an initial state, to state s355, which is a final target state. Using the control prediction model 15c shown in Fig. 5, one or more actions, i.e., operations, required for transition from state s190, which is the initial state, to state s355, which is a final target state, are uniquely identified, as shown in Fig. 6. The control prediction data 15d includes operations and the order of operations for the process until the target value of a physical quantity is reached in response to an input of the target value.

[0048] (Process prediction model 15e and characteristic parameters 15g according to the embodiment) 7 is a diagram showing characteristic parameters 15g and a process prediction model 15e according to the embodiment, and shows an example of the process prediction model 15e and the relationship between the process prediction model 15e and the characteristic parameters 15g.

[0049] The process prediction model 15e is expressed as a model constructed by a predetermined theoretical formula. In FIG. 7, the predetermined theoretical formula is expressed as y=ax 2 +bx+c. In this example, the initial values ​​of the parameters a, b, and c of the process prediction model 15e are given as 1, 0, and 0, respectively.

[0050] It also shows that the correction amounts of the parameters of the process prediction model 15e are calculated as a: 0.2, b: 0, c: 0 for the product type I, and a: 0, b: -0.1, c: 0 for the equipment A. The correction amounts (correction parameters) of the parameters of the process prediction model 15e are stored in the characteristic parameters 15g for each product type and each equipment type 3.

[0051] The initial values ​​of the parameters a, b, and c of the process prediction model 15e are corrected by the correction amounts for each type of equipment and each piece of facility equipment 3, and as a result, the parameters a, b, and c of the corrected process prediction model are calculated as 1.2, −0.1, and 0, respectively.

[0052] 7 shows an example of a table in which the characteristic parameters 15g for the types and the equipment 3 are stored, and the table stores as many characteristic parameters 15g as the number of characteristics. Also, in FIG. 7, the process prediction model 15e is 2 Since the model is expressed as +bx+c, the description is based on the assumption that the parameters are composed of three elements: a, b, and c. However, as described above, when the process prediction model 15e is expressed in other ways, the number of characteristic parameters stored corresponds to the expression format.

[0053] The process prediction model 15e is trained as follows. First, the processor 14 acquires operation record data 15a for each characteristic of the equipment 3, which relates to the operation record of the equipment 3, including measured values ​​of physical quantities in the process. Next, the processor 14 acquires the characteristics from the operation record data 15a. Next, the processor 14 calculates, for each characteristic, correction parameters for correcting the process prediction model 15e so that the actual values ​​of the physical quantities of the process executed by the equipment 3 and the estimated values ​​of the physical quantities estimated by the process prediction model 15e satisfy a predetermined condition. The predetermined condition is, for example, a condition that the difference between the actual measured quantity and the estimated quantity of the process prediction model is minimized. Next, the processor 14 corrects the process prediction model 15e using the correction parameters for each characteristic. The process prediction model 15e corrected in this way becomes the created process prediction model 15e.

[0054] (Prediction process according to the embodiment) FIG. 8 is a diagram illustrating a prediction process according to the embodiment.

[0055] First, in step S11, the data collection unit 14a acquires the operation performance data 15a corresponding to a predetermined period from the operation performance data 15a accumulated in the storage unit 15 for each process to be controlled.

[0056] Next, in step S12, the control prediction unit 14c acquires an optimal control flow in the control prediction model 15c toward the target value of the controlled process based on the actual operation data 15a for the predetermined period acquired in step S11 and the control prediction model 15c. Specifically, the control prediction unit 14c uses the actual operation data 15a corresponding to the predetermined period acquired in step S11 as measured values ​​of physical quantities for the controlled process and inputs them into the control prediction model 15c together with future target values ​​of the physical quantities. The control prediction unit 14c then acquires the optimal control flow (control value, control set value, SV value) for the controlled process output from the control prediction model 15c. The control flow includes operations for the process and the order of operations, and is data in which the control value, control set value, SV value, etc. are arranged in chronological order in the order of execution.

[0057] Next, in step S13, the control prediction unit 14c generates control prediction data 15d (referred to as first control data) of the process to be controlled based on the control flow acquired in step S12, and stores it in the storage unit 15.

[0058] Next, in step S14, the data collection unit 14a acquires the first control data of the controlled process stored in the storage unit 15 in step S13 and the operating performance data 15a of the controlled process corresponding to a predetermined period.

[0059] Next, in step S15, the process prediction unit 14d generates prediction information of a change in state of the process (process prediction data 15f (referred to as first prediction data)), including a prediction of a change in a physical quantity when the process to be controlled is controlled based on the first control data. The process prediction unit 14d generates the first prediction data based on the first control data acquired in step S14, the operating performance data 15a corresponding to a predetermined period of the process to be controlled, and the process prediction model 15e. Then, the process prediction unit 14d stores the generated first prediction data in the storage unit 15.

[0060] Next, in step S16, the model evaluation unit 14e acquires the first control data generated in step S13 and the first prediction data generated in step S15, and outputs them from the output device 13.

[0061] (Use Case 1 (Real-time Processing) of Prediction Processing According to the Embodiment) Fig. 9 is a diagram showing use case 1 (real-time processing) of the prediction processing according to the embodiment. Use case 1 (real-time processing) is one of the use cases of the prediction processing shown in Fig. 8. In use case 1 (real-time processing), guidance regarding plant control is displayed, for example, in real time while the plant is operating, and instructions to correct the plant control are input and reflected in the plant control.

[0062] The control device 2 outputs a control instruction to the facility equipment 3 (S101) and collects various physical quantities (PV values) from the facility equipment 3 (S102). The data collection unit 14a stores the information collected by the control device 2 in the memory unit 15 as operation performance data 15a (S103).

[0063] The control prediction unit 14c outputs control prediction data 15d based on predetermined information in the operation performance data 15a stored in the storage unit 15 and the control prediction model 15c (S104). The process prediction unit 14d outputs process prediction data 15f based on predetermined information in the operation performance data 15a stored in the storage unit 15, the control prediction data 15d, and the process prediction model 15e (S105).

[0064] The control prediction data 15d output by the control prediction unit 14c and the process prediction data 15f output by the process prediction unit 14d are displayed on an administrator screen 17 output from the output device 13 (S106). The plant administrator refers to the display on the administrator screen 17 and inputs a correction instruction related to the control of the equipment 3 from the input device 12 to the control device 2 (S107). The control device 2 controls the equipment 3 based on the control prediction data 15d corrected based on the information related to the correction made by the administrator (S108).

[0065] (GUI13D1 according to use case 1 of the embodiment) FIG. 10 is a diagram showing a GUI (Graphical User Interface) 13D1 (real-time processing) related to Use Case 1 of the embodiment. The GUI 13D1 includes a selection field 131, an optimum control screen 132, and a state prediction screen 133. The selection field 131 includes a product type selection field 131a, a lot selection field 131b, and an equipment selection field 131c. A product type is selected in the product type selection field 131a. A product lot number is selected in the lot selection field 131b. An equipment 3 is selected in the equipment selection field 131c. When the product type, product lot number, and equipment 3 are selected and the execute button 131d is pressed, information related to the optimum control and the predicted state is associated and displayed on the optimum control screen 132 and the state prediction screen 133. When the correct button 131e is pressed, corrections to the product type, product lot number, and equipment 3 selection are accepted.

[0066] The optimal control screen 132 shows the actual values ​​of the past optimal control flow for the selected product type, product lot number, and raw material input amount for the equipment 3 as a solid line, and the future optimal control flow predicted based on the control prediction model 15c as a dashed line. In the future optimal control flow, the raw material input amount is changed from 20 to 15 Nm at 12:30 on 4 / 5. 3 / h, 4 / 5 13:30 15 → 10Nm 3 / h. Then, at 15:10 on April 5th, operation of Equipment 3 will cease.

[0067] The state prediction screen 133 shows, with solid lines, past states and with dashed lines, future states predicted based on the process prediction model 15e for the product conversion rate and temperature for the selected product type, product lot number, and equipment 3. In the future state, the product conversion rate will reach the termination condition at 15:10 on 4 / 5, and operation of the equipment 3 will end.

[0068] It is also possible to generate and display multiple combinations of the optimal control flow (control prediction data) shown by a dashed line on the optimal control screen 132 and the future prediction value (process prediction data) shown by a dashed line on the state prediction screen 133 as candidates. The manager can feed back the optimal control flow that he / she determines to be optimal from the multiple candidates to the facility device 3 via the control device 2.

[0069] (GUI13D1 according to use case 1 of the embodiment) Fig. 11 is a diagram showing GUI13D1 (real-time processing: comparison with initial setting flow) according to use case 1 of the embodiment. Compared to Fig. 10, Fig. 11 further shows, by dashed dotted lines, the control flow when the process is controlled according to the initial setting flow on the optimization control screen 132, and further shows, by dashed dotted lines, the transitions in product conversion rate and temperature when the process is controlled according to the initial setting flow on the state prediction screen 133.

[0070] Referring to FIG. 11, when the optimal control flow predicted based on the control prediction model 15c is implemented instead of the initial setting flow, the product conversion rate increases, the termination condition is reached quickly, and the temperature is maintained at a higher level, which shows that implementing the optimal control flow is significant.

[0071] (Use Case 2 of the Embodiment (Simulation Processing)) Fig. 12 is a diagram showing use case 2 (simulation processing) of the embodiment. Use case 2 (simulation processing) is one of the use cases of the prediction processing shown in Fig. 8. In use case 2 (simulation processing), guidance regarding a simulation of plant control under virtual conditions is displayed, and an instruction to correct the virtual conditions is input and reflected in the simulation.

[0072] Use case 2 differs from use case 1 in that the administrator, while referring to the display on the administrator screen 17, inputs a modification to the hypothetical conditions that are the premise of the simulation from the input device 12 (S107'), rather than an instruction to modify the control of the facility device 3. The modification to the hypothetical conditions is reflected in the control prediction process by the control prediction unit 14c.

[0073] (GUI13D1 according to use case 2 of the embodiment) Fig. 13 is a diagram showing a GUI13D1 (simulation) related to use case 2 of the embodiment. Compared to Fig. 10, Fig. 13 further shows, by dashed dotted lines, the control flow when the process is controlled according to the simulation results on an optimum control screen 132. Also, compared to Fig. 10, Fig. 13 further shows, by dashed dotted lines, the transitions in product conversion rate and temperature when the process is controlled according to the simulation results on a state prediction screen 133.

[0074] (Use Case 3 of the embodiment (automatic control)) Fig. 14 is a diagram showing use case 3 (automatic control) of the embodiment. Use case 3 (automatic control) is one of the use cases of the prediction process shown in Fig. 8. In use case 3 (automatic control), the control prediction data 15d is fed back and input to the control device 2, and the plant is automatically controlled. At the same time, guidance regarding the control of the plant is displayed, for example, in real time, and instructions to correct the control of the plant are input based on this display (S107) and reflected in the control of the plant.

[0075] In use case 3, unlike use case 1, correction instructions regarding the control of the equipment 3 are input from the input device 12 while referring to the display on the administrator screen 17 (S107), and the control prediction data 15d is fed back to the control device 2, and the plant is automatically controlled (S109).

[0076] (Control and monitoring process according to the embodiment) 15 is a diagram showing the control and monitoring process according to the embodiment. Step S21 of the control and monitoring process (FIG. 10) corresponds to step S11 of the prediction process (FIG. 9). Step S22 of the control and monitoring process (FIG. 10) corresponds to steps S12 to S15 of the prediction process (FIG. 9).

[0077] First, in step S21, the model creation / update unit 14f selects learning data from the driving performance data 15a for each characteristic. The selection condition for the learning data here is, for example, that the learning data has a timestamp from a recent certain period, but is not limited to this.

[0078] Next, in step S22, the model creation / update unit 14f learns the learning data selected in step S21 and creates a control prediction model 15c and a process prediction model 15e for each characteristic. Next, in step S23, the control prediction unit 14c and the process prediction unit 14d start operating the control prediction model 15c and the process prediction model 15e for each characteristic created in step S22.

[0079] The next steps S24 to S27 are a model accuracy monitoring process.

[0080] First, in step S24, the data collection unit 14a collects operation result data 15a for each characteristic and stores it in the storage unit 15. Next, in step S25, the control prediction unit 14c and the process prediction unit 14d use the control prediction model 15c and the process prediction model 15e (operating model) that were put into operation in step S23 to make predictions based on the current operation result data 15a collected in step S24. Then, the control prediction unit 14c and the process prediction unit 14d generate control prediction data 15d and process prediction data 15f.

[0081] Next, in step S26, the model evaluation unit 14e compares the predicted values ​​(control prediction data 15d and process prediction data 15f) generated in step S25 with the actual values ​​based on the current operating performance data 15a collected in step S24.

[0082] Next, in step S27, the model evaluation unit 14e determines whether the error between the predicted value and the actual value compared in step S26 is equal to or greater than a threshold. Specifically, if at least one of the error between the control prediction data 15d and its actual value and the error between the process prediction data 15f and its actual value is equal to or greater than a threshold (step S27 YES), the model evaluation unit 14e proceeds to step S28. On the other hand, if both the error between the control prediction data 15d and its actual value and the error between the process prediction data 15f and its actual value are less than the threshold (step S27 NO), the model evaluation unit 14e returns the process to step S24.

[0083] The next steps S28 to S31 are a model re-creation process.

[0084] First, in step S28, the model creation / update unit 14f selects learning data for model re-creation from the operating record data 15a for each characteristic. Here, by comparing with the selection conditions for learning data in step S21, the model creation / update unit 14f selects learning data that can reconstruct a control prediction model and a process prediction model that can perform the most accurate prediction, for example, by matching more closely with the latest data or characteristic information.

[0085] Next, in step S29, the model creation and updating unit 14f learns the learning data selected in step S28 and recreates the control prediction model 15c and / or the process prediction model 15e for each characteristic. In step S29, the model creation and updating unit 14f recreates the control prediction model 15c and / or the process prediction model 15e for which it was determined in step S27 that the error between the predicted value and the actual value is equal to or greater than the threshold.

[0086] Next, in step S30, the model evaluation unit 14e compares the model accuracy of the recreated model recreated in step S29 with that of the model in operation. That is, the model accuracy, such as the error between the predicted value and the actual value, is compared for the control prediction model 15c and / or the process prediction model 15e recreated in step S29.

[0087] Next, in step S31, the model evaluation unit 14e determines whether the recreated model has higher model accuracy than the currently operating model. If the recreated model has higher model accuracy than the currently operating model (step S31 YES), the model evaluation unit 14e proceeds to step S32. On the other hand, if the recreated model has the same or lower model accuracy as the currently operating model (step S31 NO), the model evaluation unit 14e proceeds to step S28.

[0088] In step S28, to which the process is transferred from step S31, the model creation / update unit 14f selects learning data for model re-creation from the driving performance data 15a for each characteristic. Here, the selection conditions are stricter, for example, with respect to timestamps and characteristics, compared to the selection conditions for learning data in the previous execution of step S21.

[0089] In step S32, the model creating / updating unit 14f updates the currently operating model with the recreated model that has been determined to have higher model accuracy than the currently operating model in step S31.

[0090] (Use Case 4 of the embodiment (Model update)) Fig. 16 is a diagram illustrating use case 4 (model update) of the embodiment. Use case 4 (model update) is one of the use cases of the control monitoring process illustrated in Fig. 15. In use case 4 (model update), the control prediction data 15d and / or the process prediction data 15f are recreated and updated according to the monitoring results of the model accuracy of the control prediction data 15d and the process prediction data 15f.

[0091] The model evaluation unit 14e monitors the model accuracy of the control prediction data 15d and the process prediction data 15f (S110). The model evaluation unit 14e displays information on the predicted values ​​and the actual measured values ​​based on the control prediction data 15d and the process prediction data 15f (S111) on the administrator screen 17 output from the output device 13.

[0092] When a discrepancy occurs between a predicted value based on the control prediction data 15d and / or the process prediction data 15f and an actual measurement value (S112), the model creation / update unit 14f recreates the target model (S112). At this time, learning data is selected from the operation result data 15a so as to improve the model accuracy of the control prediction data 15d and / or the process prediction data 15f to be recreated. The control prediction data 15d and / or the process prediction data 15f are updated by the recreated model.

[0093] The model creation / update unit 14f may pad the learning data (operation record data 15a) of the control prediction model 15c and the process prediction model 15e with the process prediction data 15f. This padding makes it possible to recreate models that improve the model accuracy even when sufficient data is not accumulated to update the control prediction model 15c and the process prediction model 15e, such as at the beginning of plant operation or when operating conditions are changed.

[0094] (Effects of the embodiment) In the above-described embodiment, the process control system 1 that manages a process executed in equipment 3 that constitutes a plant includes a processor 14 and a memory 11. The processor 14 generates control prediction data 15d related to a control flow for a controlled process based on a control prediction model 15c that outputs a control flow in response to an input of a target value of a physical quantity in the process. Here, the control flow includes operations and the order of operations for the process until the target value is reached.

[0095] Furthermore, based on the process prediction model 15e and the control prediction data 15d, the system generates process prediction data 15f relating to future predicted values ​​of physical quantities when the process to be controlled is controlled based on the control prediction data 15d. Here, the process prediction model 15e outputs future predicted values ​​of physical quantities in the process in response to the input of the control prediction data 15d. Then, the control prediction data 15d and information relating to the process prediction data are output from the output device 13 in association with each other.

[0096] Therefore, according to the embodiment, in plant operation support, it is possible to predict dynamic optimal control for a process to be controlled, and to monitor changes in the state of the process due to the predicted optimal control.

[0097] In the above-described embodiment, the target value of the physical quantity in the process changes over time. Therefore, according to the embodiment, a control flow based on the control prediction data 15d that is more appropriate in accordance with the target value of the physical quantity that changes over time can be fed back to the facility device 3.

[0098] In the above-described embodiment, the process is a batch process. Therefore, according to the embodiment, a control flow based on the control prediction data 15d that is more appropriate for following the batch process can be fed back to the facility device 3.

[0099] In the above-described embodiment, the control prediction data 15d is transmitted to the control device 2 that controls the facility equipment 3, and the control device 2 controls the facility equipment 3 based on the control prediction data 15d. Therefore, according to the embodiment, the control prediction data 15d can be fed back to the facility equipment 3.

[0100] Furthermore, in the above-described embodiment, a plurality of combinations of the control prediction data 15d and the process prediction data 15f generated corresponding to the control prediction data 15d are generated as candidates, and the candidates are output from the output device 13. Therefore, according to the embodiment, it is possible to perform manual control in which appropriate control prediction data 15d is selected from guidance that displays a plurality of combinations of the control prediction data 15d and the process prediction data 15f, and the appropriate control prediction data 15d is manually input as feedback to the facility device 3 by an administrator.

[0101] In the above-described embodiment, information related to a manager's correction of the control prediction data 15d output from the output device 13 is received, and process prediction data 15f is generated when the process to be controlled is controlled based on the control prediction data 15d corrected based on the information related to the correction. Then, the generated process prediction data 15f is output from the output device 13. Therefore, according to the embodiment, automatic control can be performed in which the control prediction data 15d is automatically corrected and fed back to the facility device 3.

[0102] In the above-described embodiment, operation record data 15a relating to the operation record of the facility equipment 3 when controlled based on the control prediction data 15d is acquired. Then, based on the process prediction data 15f and the operation record data 15a, it is determined whether or not the control prediction model 15c and the process prediction model 15e need to be updated. Therefore, according to the embodiment, the model accuracy of the control prediction model 15c and the process prediction model 15e can be monitored, and model degradation can be detected.

[0103] Furthermore, in the above-described embodiment, learning data is selected so that the recreated control prediction model, which is recreated to update the control prediction model 15c, has higher model accuracy than the control prediction model 15c and has the highest accuracy among the selection patterns of learning data. The selected learning data is then learned to create a recreated control prediction model, and the control prediction model 15c is updated using the recreated control prediction model. The process prediction model 15e is similar to the control prediction model 15c. Therefore, according to the embodiment, when model degradation of the control prediction model 15c and the process prediction model 15e is detected, a decrease in the accuracy of these models can be suppressed.

[0104] In the above-described embodiment, when the recreated control prediction model and the recreated process prediction model are created, the process prediction data 15f predicted based on the control prediction model 15c and the process prediction model 15e before the recreate are learned in addition to the learning data selected for the recreate. Thus, according to the embodiment, the learning data (operation performance data 15a) can be padded with the process prediction data 15f.

[0105] The present invention is not limited to the above-described embodiments as they are, and in the implementation stage, the components can be modified and embodied within the scope of the gist of the present invention, or multiple components disclosed in the above-described embodiments can be appropriately combined. [Explanation of symbols]

[0106] 1: process control system, 2: control device, 3: facility equipment, 11: memory, 13: output device, 14: processor, 14a: data collection unit, 14b: control instruction unit, 14c: control prediction unit, 14d: process prediction unit, 14e: model evaluation unit, 14f: model creation and update unit, 15a: operation record data, 15a1: state transition model, 15b: characteristic information, 15c: control prediction model, 15c0: state transition model, 15d: control prediction data, 15e: process prediction model, 15f: process prediction data, 15g: characteristic parameters, 15h: evaluation information

Claims

1. A process management system that manages processes executed in equipment constituting a plant, the process control system includes a processor and a memory; The processor: generating control prediction data relating to the control flow for the process to be controlled based on a control prediction model that outputs a control flow including operations for the process and an order of operations until a target value of a physical quantity in the process is reached in response to an input of the target value of the physical quantity; generating process prediction data relating to a future predicted value of the physical quantity in a case where the process to be controlled is controlled based on the control prediction data, based on a process prediction model that outputs a future predicted value of the physical quantity in the process in response to input of the control prediction data, and the control prediction data; Information relating to the control prediction data and the process prediction data is output from an output device in association with each other. A process control system comprising:

2. 2. The process control system of claim 1, The target value of the physical quantity is a value that changes over time. A process control system comprising:

3. 2. The process control system of claim 1, The process to be controlled is a batch process. A process control system comprising:

4. 2. The process control system of claim 1, The processor: transmitting the control prediction data to a control device that controls the facility equipment; The control device controls the facility equipment based on the control prediction data. A process control system comprising:

5. 2. The process control system of claim 1, The processor: generating a plurality of combinations of the control prediction data and the process prediction data generated corresponding to the control prediction data as candidates; The candidate is output from the output device. A process control system comprising:

6. 2. The process control system of claim 1, The processor: receiving information relating to the correction of the control prediction data output from the output device; generating the process prediction data when the process to be controlled is controlled based on the control prediction data corrected based on the information related to the correction; Information relating to the control prediction data and the process prediction data is output from the output device in association with each other. A process control system comprising:

7. 2. The process control system of claim 1, The processor: acquiring operation record data relating to the operation record of the facility equipment when controlled based on the control prediction data; Determining whether the control prediction model needs to be updated based on the process prediction data and the operation performance data. A process control system comprising:

8. 8. The process control system according to claim 7, The processor: selecting learning data for the recreated control prediction model from the operating record data so that the recreated control prediction model to be recreated in order to update the control prediction model has higher model accuracy than the control prediction model and has the highest accuracy among the selection patterns of learning data; The selected learning data is learned to create the recreated control prediction model; The control prediction model is updated with the recreated control prediction model. A process control system comprising:

9. 9. The process control system according to claim 8, The processor: When creating the recreated control prediction model, the process prediction data is learned in addition to the selected learning data. A process control system comprising:

10. 2. The process control system of claim 1, The processor: acquiring operation record data relating to the operation record of the facility equipment when controlled based on the control prediction data; determining whether the process prediction model needs to be updated based on the process prediction data and the operation performance data; A process control system comprising:

11. The process control system of claim 10, The processor: selecting learning data for the recreated process prediction model from the operation record data so that the recreated process prediction model to be recreated in order to update the process prediction model has higher model accuracy than the original process prediction model and has the highest accuracy among the selection patterns of learning data; The selected learning data is learned to create the recreated process prediction model; The process prediction model is updated with the recreated process prediction model. A process control system comprising:

12. The process control system of claim 11, The processor: When creating the recreated process prediction model, the process prediction data is learned in addition to the selected learning data. A process control system comprising:

13. 2. The process control system of claim 1, The processor: acquiring operational performance data relating to the operational performance of the facility equipment, including the measured values ​​of the physical quantities in the process; determining a state of the process from the measured value of the physical quantity; classifying the determined process states into a plurality of categories by a predetermined classification process; generating a state transition model including one or more states of the process that are transition destination states to which the process transitions when a certain operation is performed on the process in a certain state, and a transition probability of the transition to the state, based on the states of the process classified into the plurality of categories; generating an optimal transition route for the process state to transition from the first state to the second state based on the state transition model, thereby generating the control prediction model; A process control system comprising:

14. 2. The process control system of claim 1, The processor: acquiring operational performance data for each characteristic of the equipment, which data includes the measured values ​​of the physical quantities in the process; Acquiring the characteristics from the operational performance data; calculating, for each of the characteristics included in the operation record data, a correction parameter for correcting the process prediction model so that an actual value of the physical quantity in the process executed by the facility device and an estimated value of the physical quantity estimated by the process prediction model constructed using a predetermined theoretical formula and initial values ​​satisfy a predetermined condition; The process prediction model is created by correcting the process prediction model using the correction parameters for each of the characteristics. A process control system comprising:

15. A process management method executed by a process management system that manages processes executed in equipment constituting a plant, comprising: the process control system includes a processor and a memory; the processor: generating control prediction data relating to the control flow for the process to be controlled based on a control prediction model that outputs a control flow including operations for the process and an order of operations until a target value of a physical quantity in the process is reached in response to an input of the target value of the physical quantity; generating process prediction data relating to a future predicted value of the physical quantity in a case where the process to be controlled is controlled based on the control prediction data, based on a process prediction model that outputs a future predicted value of the physical quantity in the process in response to input of the control prediction data, and the control prediction data; The control prediction data and the process prediction data are output from an output device. A process management method comprising the steps of:

Images