Cut tobacco dryer controller performance evaluation method and system and storage medium

Abstract

The invention relates to the technical field of controller performance evaluation, in particular to a cut tobacco dryer controller performance evaluation method and system and a storage medium, and the method comprises the steps: obtaining an initial data set; establishing a digital twin model based on the initial data set; searching a controller parameter optimal value in the digital twin model by adopting a Bayesian optimization method; the controller parameter optimal value is applied to a cut-tobacco dryer controller system, and iterative optimization is carried out; and evaluating the performance of the cut-tobacco dryer controller based on the controller parameters after iterative optimization. According to the embodiment of the invention, through real-time iterative optimization and dynamic parameter correction of the digital twinborn simulation and physical system, the global optimal solution is finally output by taking the minimum evaluation function as the target, the output variance of the system is remarkably reduced, the control precision and robustness are improved, and the direction of parameter adjustment is defined at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of controller performance evaluation, and in particular to a method, system and storage medium for evaluating the performance of a tofu drying machine controller. Background Art

[0002] Modern tobacco dryers widely utilize automated control systems. Sensors monitor key parameters such as temperature, humidity, and airflow velocity in real time, and intelligent control algorithms (primarily PID control) automatically adjust drying parameters to precisely control the moisture content of the tobacco at the outlet near the set value. However, the various control loops in the dryer are interconnected and collectively affect the final moisture content of the tobacco. This makes it extremely difficult to establish a precise model linking individual PID controller parameters with the final moisture control performance. Therefore, how to scientifically evaluate the performance of various controller parameters and identify optimization strategies to improve tobacco quality have become critical challenges for the tobacco industry.

[0003] In industrial production, controllers are core components that precisely regulate and control system outputs based on set targets. They receive sensor feedback, compare and analyze it with target values, generate corresponding control signals, and then apply them to the controlled object through actuators, thereby adjusting its behavior. Controller performance evaluation metrics are crucial for measuring the system's response quality and the controller's regulatory capabilities. Compared to subjective worker evaluations based on experience, they more objectively reflect the rationality of controller parameter settings. However, existing controller performance evaluation metrics, such as those mentioned in "Control Performance Management in Industrial Automation," include conventional metrics (overshoot, decay rate, rise time), statistically based metrics (mean error, controller output standard deviation), and advanced performance metrics (Harris index, autocorrelation coefficient). While these metrics can provide a certain degree of controller performance evaluation, they lack systematicity and comprehensiveness, and fail to provide clear guidance on how to improve controller parameters for optimal performance. Summary of the Invention

[0004] One objective of the present invention is to provide a method, system, and storage medium for evaluating the performance of a tofu dryer controller. Through closed-loop experiments, key controller evaluation indicators that influence the dryer's moisture performance are accurately determined, providing clear guidance for adjusting controller performance parameters. This method can help factories monitor controller performance degradation in real time and make timely parameter adjustments, effectively improving controller performance and ensuring the stability and accuracy of moisture control in the tofu dryer.

[0005] To achieve the above objectives, an embodiment of the present invention provides a method for evaluating the performance of a tofu drying machine controller, comprising: Get the initial dataset; Building a digital twin model based on the initial data set; Using Bayesian optimization methods to find optimal values ​​of controller parameters in the digital twin model; Applying the optimal controller parameter values ​​to the tofu drying machine controller system for iterative optimization; The performance of the tofu drying machine controller is evaluated based on the iteratively optimized controller parameters.

[0006] Optionally, obtaining the initial data set includes: Conduct N groups of closed-loop experiments, set different controller parameters in each group of closed-loop experiments, and record the corresponding closed-loop system output variance, as well as the loop set value, operating variable, and controlled variable.

[0007] Optionally, establishing a digital twin model based on the initial data set includes: According to formula (1) and formula (2), a digital twin model is established. , (1) , (2) in, For the system The state vector at time t, For the system The state vector at time t, For the system The input vector at time t, For the system The output vector at time , and are process noise and observation noise, respectively. 、 、 、 、 and is the matrix of the state equation, are the parameters in each matrix; Obtain parameters according to formula (3) and formula (4) , , (3) , (4) in, is the actual observed output value, is the output value predicted by the model, is the estimated value of the state variable.

[0008] Optionally, using a Bayesian optimization method to find optimal values ​​of controller parameters in the digital twin model includes: Based on the current initial data set, a Gaussian process model is constructed; Use the acquisition function to obtain the next sampling point; Inputting the next sampling point into the digital twin model to obtain the corresponding output variance and dynamically expanding the initial data set; Determine whether the number of iterative optimizations is completed; When it is determined that the number of iterative optimizations has been completed, the optimal value of the current controller parameters is output; If it is determined that the iterative optimization times are not completed, the process returns to the step of constructing a Gaussian process model based on the current initial data set.

[0009] Optionally, based on the current initial data set, constructing a Gaussian process model includes: According to formula (5), we can obtain the prior probability distribution. , (5) in, is the observed value of the tofu drying machine performance, is the prior mean function with parameters The value of is the prior covariance matrix; Calculate the posterior probability distribution according to formula (6) to formula (8), , (6) , (7) , (8) in, The PID parameters are When , the mean value of the controller performance is, For parameters With parameters The covariance matrix of For parameters and The covariance matrix of The PID parameters are When , the variance of the controller performance is For parameters The prior variance of .

[0010] Optionally, using the acquisition function to obtain the next sampling point includes: According to formula (9), the acquisition function is obtained. , (9) in, is the upper confidence limit, is the weight of the variance; According to formula (10), the next observation point is obtained. , (10) in, The next observation point.

[0011] Optionally, applying the optimal controller parameter value to the tow dryer controller system and performing iterative optimization includes: Applying the optimal parameter values ​​to the tofu drying machine control system; Run the tofu drying machine control system to obtain new experimental data; Based on the new experimental data, updating the digital twin model; Based on the updated digital twin model, the Bayesian optimization method is used to optimize the controller parameters.

[0012] Optionally, evaluating the performance of the tofu drying machine controller based on the iteratively optimized controller parameters includes: According to formula (11), the controller evaluation index is calculated as follows: , (11) in, is the controller evaluation index, is the minimum variance, Output variance of the current controller.

[0013] On the other hand, the present invention further provides a tow dryer controller performance evaluation system, the system comprising a processor, and the processor is configured to execute any of the above methods.

[0014] In another aspect, the present invention further provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed by a processor, any of the above methods is implemented.

[0015] Beneficial effects of the present invention: The implementation method of the present invention collects the initial operating data of the system through closed-loop experiments, constructs a high-precision digital twin model, and innovatively uses the Bayesian optimization algorithm to efficiently search for the optimal control parameters in the model. Through real-time iterative optimization of digital twin simulation and physical systems, the parameters are dynamically corrected, and ultimately the global optimal solution is output with the goal of minimizing the evaluation function. This method significantly reduces the system output variance, improves control accuracy and robustness, and clarifies the direction of parameter adjustment. Compared with the traditional trial-and-error method, the optimization efficiency is improved, providing a more stable and reliable intelligent optimization solution for complex industrial systems.

[0016] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present invention, but do not constitute a limitation of the embodiments of the present invention. In the accompanying drawings: Figure 1 Flowchart of a method for evaluating performance of a tofu drying machine controller according to one embodiment of the present invention; Figure 2 Flowchart of a method for finding optimal values ​​of controller parameters according to one embodiment of the present invention; Figure 3 A flowchart of a method for iteratively optimizing controller parameters according to one embodiment of the present invention; Figure 4 is a block diagram of a closed-loop system according to one embodiment of the present invention; Figure 5 is the posterior probability distribution of the model according to one embodiment of the present invention; Figure 6 is an upper confidence limit value according to one embodiment of the present invention; Figure 7 is the updated model posterior probability distribution according to one embodiment of the present invention; Figure 8 is the posterior probability distribution of the model after 2 iterations according to one embodiment of the present invention; Figure 9 is the posterior probability distribution of the model after 5 iterations according to one embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.

[0019] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.

[0020] like Figure 1 The flowchart of the method for evaluating the performance of a tofu drying machine controller according to one embodiment of the present invention is shown. Figure 1The evaluation method may include the following steps: In step S10, an initial data set is obtained; In step S11, a digital twin model is established based on the initial data set; In step S12, the Bayesian optimization method is used to find the optimal values ​​of the controller parameters in the digital twin model; In step S13, the optimal controller parameter values ​​are applied to the tofu drying machine controller system for iterative optimization; In step S14, the performance of the tofu drying machine controller is evaluated based on the iteratively optimized controller parameters.

[0021] In this Figure 1 In the method for evaluating the performance of the tofu drying machine controller shown, step S10 is used to obtain an initial data set. The method for obtaining the initial data set in step S10 can be various forms known to those skilled in the art. In one example of the present invention, step S10 can be to conduct a closed-loop experiment to obtain initial data. Specifically, Group experiments to obtain initial training data , , are the PID controller parameters, As the objective function (evaluation criterion), in the embodiment of the present invention, the output variance of the closed-loop system under the control of the current controller is selected, and the set value, operating variable, and controlled variable data of the loop in each experiment are recorded. , . is the set value, is the controlled variable, is the operation variable.

[0022] Step S11 is used to establish a digital twin model. The method for establishing the digital twin model in step S11 can be various forms known to those skilled in the art. In one example of the present invention, step S11 can include: In step S20, a digital twin model is established according to formula (1) and formula (2). , (1) , (2) in, For the system The state vector at time t, For the system The state vector at time t, For the system The input vector at time t, For the system The output vector at time t, and are process noise and observation noise, respectively. 、 、 、 、 and is the matrix of the state equation, are the parameters in each matrix; In step S21, the parameters are obtained according to formula (3) and formula (4): , , (3) , (4) in, is the actual observed output value, is the output value predicted by the model, is the estimated value of the state variable.

[0023] In step S22, according to formula (12) to formula (17), the Kalman filter is used to obtain the estimated value of the state variable , , (12) , (13) , (14) , (15) , (16) , (17) in, For the moment For the moment The predicted value of the state, for The estimated state value at time t, for The control input vector at time , For the moment For the moment The prediction error covariance matrix of the state, for The estimated error covariance matrix at time , For the moment For the moment The predicted value of the output, For the moment For the moment The output prediction error covariance matrix, for The optimal state estimate of is the inverse of the observed residual covariance, for The actual observed value of For the moment The state estimation error covariance matrix is ​​.

[0024] Steps S20 to S22 are used to perform parameter identification on the state space model using the prediction error method and the Kalman filter, and to establish a digital twin object for further system analysis, optimization, and control strategy design.

[0025] Step S12 is used to find the optimal value of the controller parameter in the digital twin model using the Bayesian optimization method. In this embodiment, the specific method for finding the optimal value of the controller parameter in step S12 can be various forms known to those skilled in the art. In one example of the present invention, step S12 may include: Figure 2 The steps shown in Figure 2 In the step S12, the following steps may be included: In step S30, a Gaussian process model is constructed based on the current initial data set; In step S31, the next sampling point is obtained using an acquisition function; In step S32, the next sampling point is input into the digital twin model to obtain the corresponding output variance and dynamically expand the initial data set; In step S33, it is determined whether the number of iterative optimizations is completed; In step S34, when it is determined that the iterative optimization times have been completed, the optimal value of the current controller parameter is output; If it is determined that the iterative optimization times are not completed, the process returns to the step of constructing a Gaussian process model based on the current initial data set.

[0026] In this Figure 2 In the method shown, step S30 is used to construct a Gaussian process model. Specifically, in this example, the specific method of constructing the Gaussian process model may include: In step S40, the prior probability distribution is obtained according to formula (5): , (5) in, is the observed value of the tofu drying machine performance, is the prior mean function with parameters The value of is the prior covariance matrix; In step S41, the posterior probability distribution is calculated according to formula (6) to formula (8), , (6) , (7) , (8) in, The PID parameters are When , the mean value of the controller performance is, For parameters With parameters The covariance matrix of For parameters and The covariance matrix of The PID parameters are When , the variance of the controller performance is For parameters The prior variance of .

[0027] Steps S40 and S41 use the initial data to establish a Gaussian process model for the relationship between the closed-loop system output variance and the controller parameters. This model provides estimates and uncertainties for any point. By fitting the Gaussian process to the existing observed data, a preliminary estimate of the function that relates the closed-loop system output variance to the controller parameters is created.

[0028] Step S31 is used to obtain the next sampling point using an acquisition function. In this embodiment, the specific method for obtaining the next sampling point in step S31 can be various forms known to those skilled in the art. In one example of the present invention, an upper confidence limit can be used as the acquisition function. Specifically, step S31 may include: In step S50, the acquisition function is obtained according to formula (9): , (9) in, is the upper confidence limit, is the weight of the variance; In step S51, the next observation point is obtained according to formula (10): , (10) in, The next observation point.

[0029] Steps S50 and S51 are used to obtain the next sampling point. The acquisition function will strike a balance between exploration and utilization, that is, it must perform precise optimization around existing data points and explore unknown areas. The point to be observed is the position that maximizes the acquisition function.

[0030] Step S32 is used to dynamically expand the initial data set. Specifically, in this example, the specific method of dynamically expanding the initial data set may include: after finding the next sampling point through the selected acquisition function, performing corresponding experiments, that is, adjusting the controller parameters to the sampling point And collect closed-loop system output variance data , add it to the existing dataset and update the dataset The model's posterior probability distribution is updated. At this point, the model updates its approximation of the objective function based on the new observations. The optimization continues iteratively, continuously selecting new observations and updating the model. Through this process, Bayesian optimization gradually converges to the optimal solution for the objective function.

[0031] Step S33 is used to determine whether the iterative optimization times have been completed. Step S34 is used to perform corresponding steps based on the determination result. If it is determined that the iterative optimization times have been completed, the optimal values ​​of the current controller parameters are output; if it is determined that the iterative optimization times have not been completed, the process returns to the step of constructing a Gaussian process model based on the current initial data set.

[0032] Step S13 is used to apply the optimal controller parameter value to the tow-bread drying machine controller system for iterative optimization. In this embodiment, the specific method for iterative optimization of step S13 in the actual controller system can be various forms known to those skilled in the art. In one example of the present invention, step S13 may include the following: Figure 3 The steps shown in Figure 3 In the step S13, the following steps may be performed: In step S60, the optimal parameter values ​​are applied to the tofu drying machine control system; In step S61, the tofu drying machine control system is operated to obtain new experimental data; In step S62, the digital twin model is updated based on the new experimental data; In step S63, the controller parameters are optimized using the Bayesian optimization method based on the updated digital twin model.

[0033] In this Figure 3 In the method shown, steps S60 to S63 apply the found optimal parameters to the actual control system, and then repeat steps S11 and S12 to further improve the model and optimization parameters. By continuously updating the model and optimization parameters, the performance of the control system can be gradually improved.

[0034] Step S14 is used to evaluate the performance of the tofu drying machine controller. In this embodiment, the specific method for evaluating the performance of the tofu drying machine controller in step S14 can be various forms known to those skilled in the art. In one example of the present invention, step S14 can include: According to formula (11), the controller evaluation index is calculated as follows: , (11) in, is the controller evaluation index, is the minimum variance, Output variance of the current controller.

[0035] In one embodiment of the present invention, the controller performance evaluation method described above was tested. The details are as follows: (1) Closed-loop system structure: The closed-loop system block diagram is as follows Figure 4 As shown in this Figure 4 middle, is the error channel transfer function, which in this example is , is the process channel transfer function, in this example . Pass a function to the controller, is a white noise with a variance of 1. In this example, the PI controller is selected, and the controller integral term Select 0.1, the proportional coefficient is the parameter to be adjusted, and the transfer function is Currently awaiting evaluation is 0.6. (2) Obtaining experimental data: In the interval, select a point every 0.5 (i.e. 0.1, 0.6, 1.1). value, run the closed-loop system, and measure the variance of the output signal. Get the experimental result data , (3) Experimental steps: Use the initial data to build a digital twin model and use the Bayesian method to optimize it, such as Figure 5 The posterior probability distribution of the model shown, where line b is the mean of the model The other two lines are . Select the upper confidence limit (UCB) as the acquisition function, such as Figure 6 Shown are upper confidence limits. , according to the next sampling point in this embodiment , the closed-loop system output variance data is , get the updated dataset , the posterior probability distribution of the updated model is as follows Figure 7 Then iterative optimization is performed. In this embodiment, 5 iterations are performed. As the iterations proceed, the model gradually converges to the objective function. Figure 8 For the model with 2 iterations, Figure 9 is a model that iterates 5 times. (4) Experimental results: In this embodiment , , the controller performance index is .

[0036] It can be seen that the embodiments of the present invention dynamically correct parameters through real-time iterative optimization of digital twin simulation and physical systems, output the global optimal solution, significantly reduce the system output variance, and improve control accuracy and robustness.

[0037] On the other hand, the present invention also provides a towheat drying machine controller performance evaluation system, the controller performance evaluation system includes a processor, the processor is configured to perform the following Figure 1 The method comprises: In step S10, an initial data set is obtained; In step S11, a digital twin model is established based on the initial data set; In step S12, the Bayesian optimization method is used to find the optimal values ​​of the controller parameters in the digital twin model; In step S13, the optimal controller parameter values ​​are applied to the tofu drying machine controller system for iterative optimization; In step S14, the performance of the tofu drying machine controller is evaluated based on the iteratively optimized controller parameters.

[0038] In another aspect, the present invention further provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed by a processor, any of the above methods is implemented.

[0039] Beneficial effects of the present invention: The present invention collects the system's initial operating data through closed-loop experiments, constructs a high-precision digital twin model, and innovatively uses the Bayesian optimization algorithm to efficiently search for optimal control parameters in the model. Through real-time iterative optimization of digital twin simulation and physical systems, parameters are dynamically corrected, and ultimately the global optimal solution is output with the goal of minimizing the evaluation function. This method significantly reduces the system output variance, improves control accuracy and robustness, and clarifies the direction of parameter adjustment. Compared with traditional trial-and-error methods, optimization efficiency is improved, providing a more stable and reliable intelligent optimization solution for complex industrial systems.

[0040] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0041] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0042] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0043] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0044] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0045] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0046] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0047] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0048] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for evaluating the performance of a tofu drying machine controller, characterized in that: The method comprises: Get the initial dataset; Building a digital twin model based on the initial data set; Using Bayesian optimization methods to find optimal values ​​of controller parameters in the digital twin model; Applying the optimal controller parameter values ​​to the tofu drying machine controller system for iterative optimization; The performance of the tofu drying machine controller is evaluated based on the iteratively optimized controller parameters.

2. The evaluation method according to claim 1, wherein Obtaining the initial dataset includes: Conduct N groups of closed-loop experiments, set different controller parameters in each group of closed-loop experiments, and record the corresponding closed-loop system output variance, as well as the loop set value, operating variable, and controlled variable.

3. The evaluation method according to claim 1, wherein: Based on the initial data set, establishing a digital twin model includes: According to formula (1) and formula (2), a digital twin model is established. ，（1） ，（2） in, For the system The state vector at time t, For the system The state vector at time t, For the system The input vector at time t, For the system The output vector at time , and are process noise and observation noise, respectively. 、 、 、 、 and is the matrix of the state equation, are the parameters in each matrix; Obtain parameters according to formula (3) and formula (4) , ，（3） ，（4） in, is the actual observed output value, is the output value predicted by the model, is the estimated value of the state variable.

4. The evaluation method according to claim 1, wherein: The Bayesian optimization method is used to find the optimal values ​​of controller parameters in the digital twin model, including: Based on the current initial data set, a Gaussian process model is constructed; Use the acquisition function to obtain the next sampling point; Inputting the next sampling point into the digital twin model to obtain the corresponding output variance and dynamically expanding the initial data set; Determine whether the number of iterative optimizations is completed; When it is determined that the number of iterative optimizations has been completed, the optimal value of the current controller parameters is output; If it is determined that the iterative optimization times are not completed, the process returns to the step of constructing a Gaussian process model based on the current initial data set.

5. The evaluation method according to claim 4, wherein: Based on the current initial data set, building a Gaussian process model includes: According to formula (5), we can obtain the prior probability distribution. ，（5） in, is the observed value of the tofu drying machine performance, is the prior mean function with parameters The value of is the prior covariance matrix; Calculate the posterior probability distribution according to formula (6) to formula (8), ，（6） ，（7） ，（8） in, The PID parameters are When , the mean value of the controller performance is, For parameters With parameters The covariance matrix of For parameters and The covariance matrix of The PID parameters are When , the variance of the controller performance is For parameters The prior variance of .

6. The evaluation method according to claim 4, wherein: Using the acquisition function to obtain the next sampling point includes: According to formula (9), the acquisition function is obtained. ，（9） in, is the upper confidence limit, is the weight of the variance; According to formula (10), the next observation point is obtained. ，（10） in, The next observation point.

7. The evaluation method according to claim 1, wherein: Applying the optimal controller parameter values ​​to the tofu drying machine controller system, and performing iterative optimization includes: Applying the optimal parameter values ​​to the tofu drying machine control system; Run the tofu drying machine control system to obtain new experimental data; Based on the new experimental data, updating the digital twin model; Based on the updated digital twin model, the Bayesian optimization method is used to optimize the controller parameters.

8. The evaluation method according to claim 1, wherein: Based on the iteratively optimized controller parameters, the performance evaluation of the tofu drying machine controller includes: According to formula (11), the controller evaluation index is calculated as follows: ，（11） in, is the controller evaluation index, is the minimum variance, Output variance of the current controller.

9. A tofu drying machine controller performance evaluation system, characterized in that: The system comprises a processor configured to execute the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images