Process control method and apparatus

Abstract

The present application provides a process control method, which comprises: a PID controller outputs a control value to a plurality of devices in a process, the plurality of devices in the process generates an output value according to the control value; a process simulation model corresponding to the process is updated according to the control value and the output value; parameters of the PID controller are optimized according to the updated process simulation model, and the plurality of devices in the process are controlled according to the optimized parameters of the PID controller.

Description

Technical Field

[0001] This invention relates primarily to the field of process control, and more particularly to a process control method and apparatus. Background Technology

[0002] In industrial control, PID controllers dominate. The control performance of a PID controller relies on fine-tuning of its parameters (proportional, integral, and derivative). PID control typically involves independent control loops, but if multiple PID control loops exist, they can interact, leading to significant complexity in global PID control. Based on this, Model Predictive Controllers (MPCs) have been introduced to determine optimal controller parameters based on process models. However, model drift can have unpredictable and catastrophic consequences. Furthermore, production lines may face various production scenarios, such as different input flow rates and quality requirements, which existing PID controllers cannot adapt to. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a process control method and apparatus to dynamically adjust the control parameters of a PID controller, thereby improving the accuracy of process control and achieving global optimization of process control.

[0004] To achieve the above objectives, this invention proposes a process control method, comprising: a PID controller outputting a control value to multiple devices in a process; the multiple devices generating an output value based on the control value; updating a process simulation model corresponding to the process based on the control value and the output value; optimizing the parameters of the PID controller based on the updated process simulation model; and performing process control on the multiple devices in the process based on the optimized parameters of the PID controller. Therefore, the process simulation model is dynamically updated, and the control parameters of the PID controller are adjusted accordingly, which can eliminate the offset error of the process simulation model, avoid the process getting trapped in local optima, and achieve global optimization of the process.

[0005] Preferably, updating the process simulation model corresponding to the process based on the control value and the output value includes: training a reinforcement learning model based on the control value and the output value, and using the reinforcement learning model to update the process simulation model. Therefore, updating the process simulation model using a reinforcement learning model can improve the intelligence of the process simulation model update.

[0006] Preferably, the process control method includes: determining a nominal model and a noise model in the process simulation model, and updating the noise model in the process simulation model using the reinforcement learning model. This can improve the efficiency of updating the process simulation model.

[0007] Preferably, optimizing the PID controller parameters based on the updated process simulation model includes: configuring multiple preset process scenarios, generating an optimized parameter set corresponding to the multiple preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimized parameters in the optimized parameter set corresponding to the current process scenario. Therefore, by configuring different preset process scenarios, the accuracy of process control can be improved.

[0008] Preferably, the process control method includes: real-time detection of the process scenario, and updating optimization parameters from the optimization parameter set when the process scenario changes. Therefore, by dynamically updating the optimization parameters according to the process scenario, the real-time performance of process control can be improved.

[0009] This invention proposes a process control device, comprising: a data acquisition module, wherein a PID controller outputs control values ​​to multiple devices in the process, and the multiple devices in the process generate an output value based on the control values; an update module, wherein the process simulation model corresponding to the process is updated based on the control values ​​and the output values; and an optimization module, wherein the parameters of the PID controller are optimized based on the updated process simulation model, and the multiple devices in the process are controlled based on the optimized parameters of the PID controller.

[0010] Preferably, the process simulation model corresponding to the process, updated by the update module according to the control value and the output value, includes: training a reinforcement learning model according to the control value and the output value, and using the reinforcement learning model to update the process simulation model.

[0011] Preferably, the process control device includes: determining a nominal model and a noise model in the process simulation model, and updating the noise model in the process simulation model using the reinforcement learning model.

[0012] Preferably, the optimization module optimizes the parameters of the PID controller according to the updated process simulation model by: configuring multiple preset process scenarios, generating an optimization parameter set corresponding to the multiple preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimization parameters in the optimization parameter set corresponding to the current process scenario.

[0013] Preferably, the process control device includes: real-time detection of the process scenario, and updating the optimization parameters from the set of optimization parameters when the process scenario changes.

[0014] The present invention proposes an electronic device including a processor, a memory and instructions stored in the memory, wherein the instructions, when executed by the processor, implement the method described above.

[0015] The present invention proposes a computer-readable storage medium having computer instructions stored thereon, which, when executed, perform the method described above. Attached Figure Description

[0016] The following figures are intended only to illustrate and explain the invention and do not limit the scope of the invention.

[0017] Figure 1 This is a flowchart of a motion control method according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of a motion control method according to an embodiment of the present invention;

[0019] Figure 3 This is an offline / online schematic diagram of a motion control method according to an embodiment of the present invention;

[0020] Figure 4 This is a schematic diagram of a motion control device according to an embodiment of the present invention;

[0021] Figure 5 This is a schematic diagram of an electronic device according to an embodiment of the present invention.

[0022] Explanation of reference numerals in the attached figures

[0023] 100 Process Control Methods

[0024] Steps 110-130

[0025] 210 PID controller

[0026] 220 process

[0027] 230 Model Calibration Module

[0028] 240 Process Simulation Model

[0029] 250 Optimization Modules

[0030] 260 Optimization Parameter Set

[0031] 270 Optimization Parameters

[0032] 280 Scene Configuration Module

[0033] A edge device

[0034] 410 Data Acquisition Module

[0035] 420 Update Module

[0036] 430 Optimization Module

[0037] 500 electronic devices

[0038] 510 processor

[0039] 520 memory Detailed Implementation

[0040] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described with reference to the accompanying drawings.

[0041] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the invention is not limited to the specific embodiments disclosed below.

[0042] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0043] This invention proposes a process control method. Figure 1 This is a flowchart of a process control method 100 according to an embodiment of the present invention. Figure 1 As shown, the process control method 100 includes:

[0044] Step 110: The PID controller outputs a control value to multiple devices in the process, and the multiple devices in the process generate an output value based on the control value.

[0045] The process involves multiple interconnected devices, forming the process itself. A PID controller outputs control values ​​to these devices, controlling multiple loops. Each loop has its own PID controller. For example, the process might include a first reaction tank and a second reaction tank connected by valves. The valves controlling the liquid level in the first reaction tank constitute the first control loop, and the valves controlling the liquid level in the second reaction tank constitute the second control loop. The first control loop is controlled by a first PID controller, and the second control loop is controlled by a second PID controller. Each device in the process generates an output value based on the control value; for example, the liquid level sensor in the first reaction tank outputs a first liquid level output value, and the liquid level sensor in the second reaction tank outputs a second liquid level output value. In some cases, the first reaction tank requires a faster flow rate, necessitating an increase in the valve's flow rate. However, increasing the valve's flow rate can cause drastic changes in the liquid level in the second reaction tank, preventing it from reaching the required level and thus leading to a local optimum.

[0046] Figure 2 This is a schematic diagram of a motion control method according to an embodiment of the present invention. Figure 2 As shown, the PID controller 210 controls the process 220. It can be understood that the PID controller 210 includes multiple sub-PID controllers, each corresponding to a different PID control loop. The process 220 includes multiple devices and the connection relationships between them.

[0047] Step 120: Update the process simulation model corresponding to the process based on the control value and output value.

[0048] A process simulation model is a simulation model established after simulating a process. It can reflect the process's topology and dynamic changes. After the PID controller outputs control values ​​to the process, the process outputs an output value based on the control values. The control values ​​and output values ​​are used to update the corresponding process simulation model. In some embodiments, updating the corresponding process simulation model based on the control values ​​and output values ​​includes: training a reinforcement learning model based on the control values ​​and output values, and using the reinforcement learning model to update the process simulation model. In some embodiments, a nominal model and a noise model are determined in the process simulation model, and the noise model in the process simulation model is updated using the reinforcement learning model.

[0049] Continue to refer to Figure 2As shown, the control value output by the PID controller 210 and the output value output by the process 220 are sent to the model calibration module 230. The model calibration module 230 can be configured with a reinforcement learning model. For this reinforcement learning model, the process can be considered the environment, and the process's input and output, i.e., the control value and output value, can be considered the state. The reinforcement learning model generates an action based on the state and reward, and this action is applied to the environment. Through the reinforcement learning model, the process simulation model can be intelligently updated. The process simulation model can typically be divided into a nominal model and a noise model. The nominal model is the framework of the process simulation model and remains largely unchanged, while the noise model is the noise component of the process simulation model and usually changes. After determining the nominal and noise models in the process simulation model, the noise model in the process simulation model can be updated using the reinforcement learning model, which can improve the efficiency of updating the process simulation model. At this point, the model calibration module 230 has updated the process simulation model 240.

[0050] Step 130: Optimize the parameters of the PID controller based on the updated process simulation model, and perform process control on multiple devices in the process based on the optimized parameters of the PID controller.

[0051] After the process simulation model is updated, the parameters of the PID controller are optimized based on the updated model, and the optimized parameters of the PID controller are used to control multiple devices in the process. Therefore, the process simulation model is dynamically updated, and the control parameters of the PID controller are adjusted accordingly. This eliminates the offset error of the process simulation model, avoids the process getting trapped in local optima, and enables the achievement of global optima for the process.

[0052] In some embodiments, optimizing the parameters of the PID controller based on the updated process simulation model includes: configuring multiple preset process scenarios, generating multiple sets of optimized parameters corresponding to the preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimized parameters in the set of optimized parameters corresponding to the current process scenario. In some embodiments, the process control method includes: real-time detection of the process scenario, and updating the optimized parameters from the set of optimized parameters when the process scenario changes.

[0053] exist Figure 2In the process, the optimization module 250 generates optimization parameters based on the updated process simulation model 240. Different preset process scenarios will generate corresponding optimization parameters, which together form an optimization parameter set 260. For example, a first preset process scenario, a second preset process scenario, and a third preset process scenario can be configured. In the first preset process scenario, the flow rate is 5-20 t / h; in the second preset process scenario, the flow rate is 20-40 t / h; and in the third preset process scenario, the flow rate is 40-50 t / h. The optimization module 250 generates corresponding first, second, and third optimization parameters for these scenarios. These first, second, and third optimization parameters are composed of multiple parameters. If the flow rate in the current process scenario is detected to be 35 t / h, the second optimization parameter corresponding to the second preset process scenario is selected. Furthermore, the process scenario of process 220 is monitored in real time. If the flow rate of process 220 changes to 45 t / h, the third optimization parameter corresponding to the third preset process scenario is selected. Figure 2 The model calibration module 230, process simulation model 240, optimization module 250, optimization parameter set 260, optimization parameter 270, and scene configuration module 280 can be fixed to hardware such as edge device A.

[0054] Figure 3 This is an offline / online schematic diagram of a motion control method according to an embodiment of the present invention. The left side of the dashed line represents the offline module, and the right side of the dashed line represents the online module, namely the process simulation module 240. The optimization module 250, the optimization parameter set 260, and the scene configuration module 280 can be pre-fixed and applied to the online module in process control.

[0055] The embodiments of the present invention provide a process control method in which the process simulation model of the process control is dynamically updated and the control parameters of the PID controller are adjusted accordingly. This can eliminate the offset error of the process simulation model, avoid the process from getting trapped in local optima, and achieve global optima of the process.

[0056] The present invention also proposes a process control device. Figure 4 This is a schematic diagram of a motion control device 400 according to an embodiment of the present invention. The process control device 400 includes:

[0057] The data acquisition module 410 and the PID controller output control values ​​to multiple devices in the process, and the multiple devices in the process generate an output value based on the control values;

[0058] Update module 420 updates the process simulation model corresponding to the process based on the control values ​​and output values;

[0059] The optimization module 430 optimizes the parameters of the PID controller based on the updated process simulation model, and performs process control on multiple devices in the process based on the optimized parameters of the PID controller.

[0060] In some embodiments, the process simulation model corresponding to the update module based on the control value and the output value includes: training a reinforcement learning model based on the control value and the output value, and using the reinforcement learning model to update the process simulation model.

[0061] In some embodiments, the process control device 400 includes: determining a nominal model and a noise model in a process simulation model, and updating the noise model in the process simulation model using a reinforcement learning model.

[0062] In some embodiments, the optimization module 430 optimizes the parameters of the PID controller according to the updated process simulation model by: configuring multiple preset process scenarios, generating a set of optimized parameters corresponding to the multiple preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimized parameters in the set of optimized parameters corresponding to the current process scenario.

[0063] In some embodiments, the process control device 400 includes: real-time detection of the process scenario, and updating optimization parameters from the set of optimization parameters when the process scenario changes.

[0064] The present invention also proposes an electronic device 400. Figure 4 This is a schematic diagram of an electronic device 400 according to an embodiment of the present invention. Figure 4 As shown, the electronic device 400 includes a processor 410 and a memory 420. The memory 420 stores instructions, which, when executed by the processor 410, implement the method 100 described above.

[0065] The present invention also proposes a computer-readable storage medium having computer instructions stored thereon, which, when executed, perform the method 100 described above.

[0066] Some aspects of the methods and apparatus of this invention can be executed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The aforementioned hardware or software may be referred to as a "data block," "module," "engine," "unit," "component," or "system." The processor may be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. Furthermore, aspects of this invention may be embodied as computer products residing in one or more computer-readable media, including computer-readable program code. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic tapes, etc.), optical discs (e.g., compact discs (CDs), digital multifunction discs (DVDs), etc.), smart cards, and flash memory devices (e.g., cards, sticks, key drives, etc.).

[0067] Flowcharts are used herein to illustrate the operations performed by the method according to embodiments of this application. It should be understood that the preceding operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more steps may be removed from them.

[0068] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0069] The above description is merely an illustrative embodiment of the present invention and is not intended to limit the scope of the invention. Any equivalent changes, modifications, and combinations made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A process control method (100), characterized in that, The process control method (100) includes: The PID controller outputs a control value to multiple devices in the process, and the multiple devices in the process generate an output value (110) based on the control value. Update the process simulation model (120) corresponding to the process based on the control value and the output value. The parameters of the PID controller are optimized based on the updated process simulation model, and process control is performed on multiple devices in the process based on the optimized parameters of the PID controller (130). The process simulation model corresponding to the process based on the control value and the output value includes: training a reinforcement learning model based on the control value and the output value, and using the reinforcement learning model to update the process simulation model; The process control method (100) includes: determining the nominal model and the noise model in the process simulation model, and updating the noise model in the process simulation model using the reinforcement learning model.

2. The process control method (100) according to claim 1, characterized in that, Optimizing the parameters of the PID controller based on the updated process simulation model includes: configuring multiple preset process scenarios, generating a set of optimized parameters corresponding to the multiple preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimized parameters in the set of optimized parameters corresponding to the current process scenario.

3. The process control method (100) according to claim 2, characterized in that, The process control method (100) includes: real-time detection of the process scenario of the process, and updating the optimization parameters from the set of optimization parameters when the process scenario changes.

4. A process control device (400), characterized in that, The process control device (400) includes: The data acquisition module (410) and the PID controller output control values ​​to multiple devices in the process, wherein the multiple devices in the process generate an output value according to the control values; The update module (420) updates the process simulation model corresponding to the process based on the control value and the output value; The optimization module (430) optimizes the parameters of the PID controller according to the updated process simulation model, and performs process control on multiple devices in the process according to the optimized parameters of the PID controller. The updating module (420) updates the process simulation model corresponding to the process according to the control value and the output value by: training a reinforcement learning model according to the control value and the output value, and using the reinforcement learning model to update the process simulation model; The process control device (400) includes: determining a nominal model and a noise model in the process simulation model, and updating the noise model in the process simulation model using the reinforcement learning model.

5. The process control device (400) according to claim 4, characterized in that, The optimization module (430) optimizes the parameters of the PID controller according to the updated process simulation model by: configuring multiple preset process scenarios, generating a set of optimization parameters corresponding to the multiple preset process scenarios, matching the current process scenario with the multiple preset process scenarios, and determining the optimization parameters in the set of optimization parameters corresponding to the current process scenario.

6. The process control device (400) according to claim 5, characterized in that, The process control device (400) includes: real-time detection of the process scenario of the process, and updating the optimization parameters from the set of optimization parameters when the process scenario changes.

7. An electronic device (500) includes a processor (510), a memory (520), and instructions stored in the memory (520), wherein the instructions, when executed by the processor (510), implement the method as claimed in any one of claims 1-3.

8. A computer-readable storage medium having stored thereon computer instructions that, when executed, perform the method according to any one of claims 1-3.

Images