Predictive control module implementation method based on domestic DCS platform

Abstract

The invention relates to a prediction control module implementation method based on a localized DCS platform, and the method comprises the following steps: developing a prediction control algorithm program, building a prediction model needed by a matrix variable expression algorithm, and inputting the algorithm program including control parameters and model update variables; storing a prediction model obtained through offline modeling or an optimization algorithm as a data file, and storing the data file in a set path; the algorithm program loads the prediction model by reading the data file under the set path, and judges the program running state according to the value of the current model update variable; when the program is in a normal operation state, control parameters are obtained, and a control instruction is output through the prediction model; when the program is in a model loading state, the latest data file under the set path is read again, the prediction model is updated, and in the updating process, the program outputs other control loop tracking instructions or manual control operation instructions. Compared with the prior art, the method has the advantages of good control performance, good flexibility and the like.

Description

Technical Field

[0001] The present invention relates to the field of power generation automatic control, and in particular to a method for realizing a prediction control module based on a domestic DCS platform. Background Art

[0002] Traditional DCS (Distributed Control System) is limited by the closed system and insufficient computing power, and can only implement conventional control methods, which cannot solve many control problems of thermal power units. Predictive control can predict the system output in advance based on the prediction model, and then plan the future control quantity, showing excellent control performance in industrial slow objects. With the research and promotion of the domestic DCS platform, the domestic DCS platform has the characteristics of independent development, flexible application and high-performance controller, which provides a platform foundation for the embedded application of predictive control algorithms. However, due to factors such as equipment aging and system transformation, the characteristics of generator equipment and systems will gradually change, and the predictive control algorithm relies on an accurate prediction model. In order to prevent the control performance from deteriorating due to model mismatch, the prediction model needs to be updated from time to time. This operation requires recompiling and downloading the controller. For the sake of safe operation, it usually needs to be performed during the unit shutdown. The inability to update the prediction model in time leads to poor actual application of the predictive control algorithm.

[0003] In traditional DCS, the only way to implement predictive control algorithms is to use PLC or server "plug-ins", which has many problems such as poor openness, difficult operation and maintenance, inconsistent standards, and chaotic systems. With the continuous demand for intelligence, many domestic DCS platform manufacturers have begun to open up system development permissions, providing a foundation for the embedded application of predictive control algorithms. However, for predictive control algorithms that rely on models, if the existing solutions need to update the model, they can only recompile and download the controller after offline modeling. For safety reasons, this operation is generally performed during unit shutdown, resulting in poor application results. Summary of the invention

[0004] The purpose of the present invention is to provide a method for implementing a prediction control module based on a domestic DCS platform. Aiming at the problems that it is difficult to update the prediction model of the prediction control module based on the domestic DCS platform and the actual application effect is not good, the present invention adopts the model file transmission method to generate a corresponding file for each prediction model, save it to an independent loading area, cooperate with the model update variable setting, flexibly load the prediction model, realize the online update of the prediction model, ensure the real-time and accuracy of the control instructions output by the prediction control module, and improve the intelligent control level of the unit.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for implementing a predictive control module based on a domestic DCS platform includes the following steps:

[0007] According to the development regulations of the domestic DCS platform, a predictive control algorithm program is developed, and the prediction model required in the algorithm is represented by matrix variables. The input of the predictive control algorithm program includes control parameters and model update variables;

[0008] The prediction model obtained by offline modeling or optimization algorithm is saved as a data file in the form of matrix variables and stored in the set path;

[0009] The predictive control algorithm program loads the prediction model by reading the matrix variables in the data file under the set path, and determines the program running state according to the value of the current model update variable, and the program running state includes the normal running state and the loading model state;

[0010] When the program is in normal operation, the control parameters are obtained and the control instructions are output using the prediction model;

[0011] When the program is in the model loading state, it re-reads the latest data file in the set path and updates the prediction model according to the matrix variables in the data file. During the updating process, the program output tracks other control loop instructions or runs manual control instructions.

[0012] The model update variable is of Boolean type. When its value is true, the prediction model needs to be updated and the program enters the model loading state. When its value is false, the prediction model does not need to be updated and the program is in normal operation.

[0013] The data file is updated in an overwrite mode. When a data file is updated, the original data file is overwritten in the corresponding path.

[0014] When the program enters the model loading state for a preset time, the model is updated to update the value of the variable so that the program enters the normal operation state, and the program outputs control instructions according to the updated prediction model.

[0015] In the domestic DCS platform development environment, the predictive control algorithm program is compiled into a dynamic link library file and packaged into a graphical predictive control module according to the packaging rules. The input of the program is packaged as the input pin in the corresponding graphical predictive control module. The module performs calculations by loading the dynamic link library file, reads the input required by the program by obtaining the value of the input pin, and outputs it through the output pin.

[0016] The graphical predictive control module reads the module ID number assigned to itself by the domestic DCS platform, and automatically establishes a corresponding folder in the controller as the set path according to the module ID number; when multiple graphical predictive control modules are added to the domestic DCS platform, the modules read the data files under the respective paths through the set paths generated by the module ID numbers.

[0017] When the method is applied to the main steam temperature control of a generator set, the control parameters include a measured value, a set value, an operation cycle, a manual state tracking value, an output upper limit, an output lower limit, a manual automatic state, a modeling time domain, a prediction time domain, a control time domain, an output error coefficient, and a control increment coefficient.

[0018] The output pin of the graphical prediction control module also includes an error reporting pin. When the program outputs an error, the error reporting pin outputs error information.

[0019] The output pins of the graphical predictive control module also include reserved pins and check pins, wherein the reserved pins are used for subsequent program expansion, and the check pins are used for checking program problems in the module.

[0020] The model update variables are set by manual tuning.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] In view of the problem that it is difficult to update the prediction model online in the prediction control module based on the domestic DCS platform, the present invention adopts the model file transmission method to save the prediction model used by each graphical prediction control module in the form of a file to an independent loading area, and combines the setting of the model update variable to flexibly load the model, thereby realizing the online update of the prediction model, so that the output control instructions can accurately and real-time track the expected operating status of the system, and provide strong support for improving the automation level of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flow chart of the method of the present invention;

[0024] Figure 2 A schematic diagram of input and output pins of a graphical predictive control module in one embodiment. DETAILED DESCRIPTION

[0025] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0026] Example 1

[0027] This embodiment provides a method for implementing a predictive control module based on a domestic DCS platform. Figure 1 As shown, the following steps are included:

[0028] S1, develop the predictive control algorithm program and represent the prediction model required by the algorithm with matrix variables.

[0029] This embodiment develops a predictive control algorithm program according to the development specifications of the domestic DCS platform, and establishes a matrix variable G as a representation of the predictive model. The program loads G by reading a data file under a set path.

[0030] A special input signal, model update variable s, is added to the predictive control algorithm program. Model update variable s is of Boolean type and takes values ​​of True or False. When it takes a value of True, the prediction model needs to be updated and the program loads the model state; when it takes a value of False, the prediction model does not need to be updated and the program is in normal operation.

[0031] In this embodiment, the model update variable is manually set according to actual needs. In other embodiments, an automatic setting method can also be adopted. For example, when a data file update is detected, the value of the model update variable is automatically changed from False to True.

[0032] That is, the input of the predictive control algorithm program in this application includes control parameters and model update variables.

[0033] S2, saves the prediction model obtained by offline modeling or optimization algorithm as a data file in the form of matrix variables and stores it in the set path.

[0034] In this embodiment, the update of the data file adopts the overwrite mode. When the data file is updated, the original data file is overwritten in the corresponding path. After the prediction model is obtained through offline training, the prediction model is saved as a data file and updated to the data file in the set path.

[0035] S3, the predictive control algorithm program loads the prediction model by reading the matrix variables in the data file under the set path, and determines the program running status according to the value of the current model update variable, and the program running status includes the normal running status and the loading model status.

[0036] In this embodiment, in the domestic DCS platform development environment, the predictive control algorithm program is compiled into a dynamic link library file, and is packaged into a graphical predictive control module according to the packaging rules. The input of the program is packaged as the input pin in the corresponding graphical predictive control module. The module performs calculations by loading the dynamic link library file, reads the input required by the program by obtaining the value of the input pin, and outputs it through the output pin.

[0037] The graphical predictive control module reads the module ID number assigned to it by the domestic DCS platform, and automatically creates a corresponding folder in the controller as the set path based on the module ID number. When multiple graphical predictive control modules are added to the domestic DCS platform, the set path generated by the module ID number reads the data files under each path to avoid confusion in reading file paths and ensure that the model parameter file uniquely corresponds to the algorithm module.

[0038] S4, when the program is in normal operation, the control parameters are obtained and the prediction model composed of the matrix variable G is used to output the control instructions.

[0039] S5, when the program is in the model loading state, re-read the latest data file under the set path, and update the prediction model according to the matrix variable G in the data file. During the updating process, the program output tracks other control loop instructions or runs manual control instructions.

[0040] In this embodiment, when the data file under the set path is updated, the input pin corresponding to the model update variable s of the graphical predictive control module is set to True, so that the program enters the model loading state.

[0041] When the program enters the loading model state for 10 seconds, the model update is considered to be completed, and the value of the model update variable is updated to False so that the program enters the normal operation state and outputs control instructions according to the updated prediction model.

[0042] After completing the online update of the prediction model, the graphical prediction control module enters the normal control process. During the online operation process, if it is found that the output control instructions are difficult to meet the current needs, the value of the model update variable s can be set to make it enter step S5 again for online update without compiling and downloading the controller.

[0043] Example 2

[0044] This embodiment takes the application in the main steam temperature control of a generator set as an example to further illustrate the method described in Embodiment 1.

[0045] The main steam temperature control of a 660MW generator set originally adopted a cascade PID control strategy. Affected by the peak load and frequency regulation of the power grid, the daily operating conditions of the unit often change greatly and rapidly, which makes the main steam temperature of the unit prone to overheating, and has a great impact on the stable and economic operation of the unit. In order to improve the main steam temperature control effect, a predictive control algorithm is implemented based on a domestic control platform. However, since the characteristics of the boiler combustion system will change with the increase of operating time, the actual operation effect of the predictive control algorithm is not good. It is necessary to update the prediction model during the unit shutdown period. The untimely update of the model makes it impossible to improve the control performance in time. If the prediction model is not updated online, it will be difficult to ensure that the main steam temperature is controlled to the desired state. Therefore, this embodiment adopts the method of Example 1 to implement the predictive control algorithm, and the implementation process is as follows:

[0046] (1) According to the development specifications of the domestic DCS platform, a predictive control algorithm program is developed, and a matrix variable G is established as the prediction model required by the predictive control algorithm program. The program loads G by reading the data file under the set path.

[0047] (2) Add a special input signal to the predictive control algorithm program - the predictive model update variable s, and perform logical judgment on s. The program needs to read the input parameters required by the predictive control algorithm: measured value PVin, set value SPin, operation cycle DT, manual state tracking value TR, output high limit HIM, output low limit LIM, manual automatic state STR, modeling time domain N, prediction time domain P, control time domain M, output error coefficient Q, control increment coefficient Lamda, and the value of the model update variable s, s takes the value of True or False. When the value of s is False, the program switches to the normal operation state, and the prediction model composed of the matrix variable G obtains the calculation result and outputs it as a control instruction; when the value of s is True, the program switches to the loading model state, outputs tracking instructions for other control loops or runs manual control instructions, and at the same time, the program reads the data file under the set path, assigns the read data to the matrix variable G, and updates the predictive model.

[0048] (3) In a domestic DCS platform development environment, the predictive control algorithm program is compiled into a dynamic link library file and packaged into a graphical predictive control module according to the packaging rules. The input and output of the program are packaged as the input and output pins in the corresponding graphical module. The module performs calculations by loading the dynamic link library file and reads the input required by the program by obtaining the value of the input pin.

[0049] (4) Add a graphical predictive control module in the configuration interface of the domestic DCS platform and connect the input and output pins. PVin is connected to the actual value of the superheated steam temperature, SPin is connected to the set value of the superheated steam temperature, DT is set to 1, TR is connected to the actual opening of the spray desuperheater valve to be tracked, HIM is set to 100, LIM is set to -100, STR is connected to the manual automatic switching command, N is set to 300, P is set to 200, M is set to 5, Q and Lamda are set to 1. The graphical predictive control module constructed in this embodiment is as follows Figure 2 As shown, it also includes an error pin Error. When the program outputs an error, the error pin outputs error information. The graphical predictive control module also includes a reserved pin and a check pin. The reserved pin Reserve is used for subsequent program expansion, and the check pin Check is used to check program problems in the module. The module ID number assigned by DCS is 1. The module creates a folder named "DATA1" in the controller as the setting path based on the ID number.

[0050] (5) After the prediction model is obtained through offline modeling or optimization algorithm calculation, the model is saved as a data file and updated to the "DATA1" folder. The input pin corresponding to the model update variable s is set to True and set to False after 10 seconds to complete the online update of the prediction model.

[0051] (6) After completing the online update of the prediction model, the system enters the normal operation state and realizes the system control process. During the online commissioning process, step (5) can be repeated to perform online update again without compiling and downloading the controller.

[0052] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for implementing a predictive control module based on a domestic DCS platform, characterized in that: The following steps are involved: Developing a predictive control algorithm program and representing the prediction model required by the algorithm with matrix variables, wherein the input of the predictive control algorithm program includes control parameters and model update variables; The prediction model obtained by offline modeling or optimization algorithm is saved as a data file in the form of matrix variables and stored in the set path; The predictive control algorithm program reads the matrix variables in the data file under the set path to load the predictive model, and determines the program running state according to the value of the current model update variable, and the program running state includes the normal running state and the loading model state; When the program is in normal operation, the control parameters are obtained and the control instructions are output using the prediction model; When the program is in the model loading state, it re-reads the latest data file in the set path and updates the prediction model according to the matrix variables in the data file. During the updating process, the program output tracks other control loop instructions or runs manual control instructions.

2. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: The model update variable is of Boolean type. When its value is true, the prediction model needs to be updated and the program enters the model loading state. When its value is false, the prediction model does not need to be updated and the program is in normal operation.

3. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: The data file is updated in an overwrite mode. When a data file is updated, the original data file is overwritten in the corresponding path.

4. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: When the program enters the model loading state for a preset time, the model is updated to update the value of the variable so that the program enters the normal operation state, and the program outputs control instructions according to the updated prediction model.

5. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: In the domestic DCS platform development environment, the predictive control algorithm program is compiled into a dynamic link library file and packaged into a graphical predictive control module according to the packaging rules. The input of the program is packaged as the input pin in the corresponding graphical predictive control module. The module performs calculations by loading the dynamic link library file, reads the input required by the program by obtaining the value of the input pin, and outputs it through the output pin.

6. A method for implementing a predictive control module based on a domestic DCS platform according to claim 5, characterized in that: The graphical predictive control module reads the module ID number assigned to itself by the domestic DCS platform, and automatically establishes a corresponding folder in the controller as the set path according to the module ID number; when multiple graphical predictive control modules are added to the domestic DCS platform, the modules read the data files under the respective paths through the set paths generated by the module ID numbers.

7. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: When the method is applied to the main steam temperature control of a generator set, the control parameters include a measured value, a set value, an operation cycle, a manual state tracking value, an output upper limit, an output lower limit, a manual automatic state, a modeling time domain, a prediction time domain, a control time domain, an output error coefficient, and a control increment coefficient.

8. The method for implementing a predictive control module based on a domestic DCS platform according to claim 5, characterized in that: The output pin of the graphical prediction control module also includes an error reporting pin. When the program outputs an error, the error reporting pin outputs error information.

9. The method for implementing a predictive control module based on a domestic DCS platform according to claim 5, characterized in that: The output pins of the graphical predictive control module also include reserved pins and check pins, wherein the reserved pins are used for subsequent program expansion, and the check pins are used for checking program problems in the module.

10. The method for implementing a predictive control module based on a domestic DCS platform according to claim 1, characterized in that: The model update variables are set by manual tuning.

Images