Coordination control system based on disturbance suppression predictive control

Abstract

The invention discloses a coordination control system based on disturbance suppression predictive control, which comprises a data acquisition unit, a wireless transmission module, a coordination control system and a controllable display interface, and relates to the technical field of thermal power regenerative control. According to the coordination control system based on disturbance suppression predictive control, the coordination control system is arranged, after data acquisition and processing are completed, training compensation of historical data is achieved through a coordination analysis model, then real-time data is introduced to achieve active coordination after a load instruction changes, and error values reflected between compensation devices are combined, so that the disturbance suppression predictive control is achieved. According to the method, a prediction point capable of being coordinated in advance and a disturbance deviation value of regenerative master control are found, synchronous coordination control processing operation is realized, the coordination efficiency is effectively improved, the stability in the coordination process is guaranteed, the error influence is effectively reduced, and the error-tolerant rate of coordination control is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power regenerative control, and specifically to a coordinated control system based on disturbance rejection predictive control. Background Art

[0002] As an important part of the power system, the operating efficiency and stability of thermal power units have an important impact on the stable operation of the power system. As an important part of thermal power units, the operating efficiency of the regenerative system is directly related to the overall performance of the units.

[0003] Refer to the patent with the name: A coordinated control method for thermal power units based on disturbance rejection predictive control (Patent Publication No.: CN114706301A, Patent Publication Date: July 5, 2022). The method includes the following steps: Step S1: Establish a prediction equation for the output controlled variable according to the prediction model; Step S2: Optimize and calculate the input control variable by using the performance index function; Step S3: Introduce a setpoint filter to perform transition processing on the reference setpoint, which can effectively improve the disturbance rejection ability of the coordinated system of thermal power units, can solve the limited optimization problems of the control action rate and amplitude in the coordinated control system of thermal power units, and the control is fast, accurate, with excellent control performance, and the unit operates stably and reliably.

[0004] Based on the description of the above document, during the operation of the existing coordinated control system, the turbine main controller responds quickly to the received instruction, while the boiler main controller has a delay in receiving the instruction. At the same time, the associated regenerative main controller operation is affected by external disturbances and internal parameter changes, resulting in low control accuracy and poor stability, so that the deviation in the coordinated operation is large, the stability is not high, and the efficiency of the coordinated operation is affected. Therefore, the present invention provides a coordinated control system based on disturbance rejection predictive control. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a coordinated control system based on disturbance rejection predictive control, which solves the problems that during the operation of the existing coordinated control system, the turbine main controller responds quickly to the received instruction, while the boiler main controller has a delay in receiving the instruction. At the same time, the associated regenerative main controller operation is affected by external disturbances and internal parameter changes, resulting in low control accuracy and poor stability, so that the deviation in the coordinated operation is large, the stability is not high, and the efficiency of the coordinated operation is affected.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A coordinated control system based on disturbance rejection predictive control, comprising:

[0007] A data acquisition unit, configured to collect the real-time operation data of the boiler and the turbine, and extract the historical operation data;

[0008] A wireless transmission module uses wireless communication technology to achieve data transmission and stores the collected data in a database;

[0009] A coordinated control system realizes data analysis and processing and conducts coordinated control according to a load instruction. Specifically, it includes:

[0010] A data analysis unit realizes the classification operation of data, forming required historical data sets and real-time data sets;

[0011] A prediction compensation unit establishes a coordinated analysis model, uses the historical data set to train the coordinated analysis model, and then introduces the real-time data set to perform a coordinated control operation after a change in the load instruction, marking the predicted control points;

[0012] An optimization transmission unit completes the synchronous coordinated control operation and compensates for the delay error by realizing data transmission operations when reaching the control points;

[0013] A controllable display interface realizes the control operation of various devices through the interface and displays the data changes during the operation.

[0014] Preferably, the operation of classifying data in the data analysis unit is as follows:

[0015] A1. First, convert the data signals collected by the sensor into digital signals that can be recognized and processed, and then summarize them into initial data;

[0016] A2. Through an extraction function, process the initial data, extract corresponding parameters from the initial data according to the required data requirements, and form the required historical data and real-time data;

[0017] A3. Classify the historical data and real-time data respectively to form the required historical data set labeled as P and real-time data set labeled as Q.

[0018] Preferably, the processing operation of the extraction function in A2 is as follows:

[0019] a21. Set an extraction function F(u, v), where u is the requirement for extracting parameters from the initial data and v is the condition for extracting parameters from the initial data;

[0020] a22. Use the subsequent required data categories as the extraction requirements of the extraction function F(u, v), and the parameter names under the subsequent required data categories as the extraction conditions of the extraction function F(u, v);

[0021] a23. Then, match the extraction requirements of the extraction function F(u, v) with the content in the initial data, and then match the extraction conditions of the extraction function F(u, v) with the same initial data that has been matched. Thus, the parameters obtained through the matching are extracted and classified to form the required data set.

[0022] Preferably, the operation of establishing a coordinated analysis model in the prediction compensation unit is as follows:

[0023] B1. Extract the corresponding device data from the historical data and extract the operation connection relationship data between the devices;

[0024] B2. Then integrate the data to form a coordinated analysis model for realizing the simulation control operations of the turbine main control, boiler main control, and regenerative heat main control. Then introduce the extracted historical data into the coordinated analysis model for training and analysis.

[0025] Preferably, the operation of the coordinated analysis model analyzing the historical data set is as follows:

[0026] C1. Extract the parameter data corresponding to different load commands as actual values, and then, during the process of the change of the load command, analyze and simulate using the coordinated analysis model;

[0027] C2. When a load change command is generated, perform the operation on the turbine main control, and analyze the influence of the coordinated control of the boiler main control after the turbine main control responds;

[0028] C3. When a load change command is generated, after completing the operation of the turbine main control, analyze the influence of the extraction amount of the regenerative heat main control on the coordinated control.

[0029] Preferably, the steps of analyzing the influence of the coordinated control of the boiler main control after the turbine main control responds in C2 are as follows:

[0030] c21. When the difference value of the load change command is J k At this time, the load change command forms a turbine load command and a boiler load command, and is transmitted to the turbine main control and the boiler main control to generate commands to control each sub-control system;

[0031] c22. When the data is introduced into the coordinated analysis model for simulation, the time data from when the turbine main control responds to the completion of the regulation operation is marked as T m Extracted, the time data from when the boiler main control responds coordinately to the completion of the regulation is T n Extracted, and during the control process, ensure the pressure balance between the front of the turbine and inside the boiler;

[0032] c23. Then, according to the deviation between the time T m of the coordination of the turbine main control and the time T n of the coordination of the boiler main control and the difference value J of the load change commandk Determine the variation coefficient of deviation regulation and label it as L i ;

[0033] c24. Subsequently, through multiple load change commands and repeating the operations of c21 to c23, multiple variation coefficients L are obtained i , and then obtain the average variation coefficient corresponding to the load change command, and calculate the regulation error time under the corresponding load change command based on the average variation coefficient, so as to determine the prediction point when the boiler master control coordinates to receive the command transmitted by the turbine master control and label it as M.

[0034] Preferably, the variation coefficient L in c23 i has the following calculation formula:

[0035] L i =|T m -T n | / J k ;

[0036] i, m, n, and k represent the corresponding category parameters of the nth one;

[0037] And the calculation formula for the average variation coefficient corresponding to c24 is: L 平 =(L1 + L2 + … + L i ) / i;

[0038] Then, based on the average variation coefficient, calculate the difference J k of the regulation error time under the corresponding load change command as: T k =L 平 ×J k ;

[0039] At this time, the time point of the prediction point M is the time when the turbine master control starts to self-regulate from the start time T k and transmits the command to achieve the coordinated response regulation of the boiler master control.

[0040] Preferably, the steps of the C3 analysis of the extraction amount of the regenerative main control on the coordinated control are as follows:

[0041] c31. According to the setting in c21 operation, when the difference of the load change command is J k , the coordinated control between the turbine master control and the boiler master control is carried out through the regenerative main control;

[0042] c32. Judge the parameter numerical result of the load command change;

[0043] Result 1. If the parameter value of the load command change becomes larger, the operation control of the regenerative main control shortens the time for coordinating between the boiler master control and the turbine master control, and the corresponding shortened time of the regenerative main control is t x ;

[0044] Result 2: If the parameter value of the load command change becomes smaller, the time for the regenerative main control to achieve coordination between the boiler main control and the steam turbine main control increases, and the corresponding increased time for the regenerative main control is t y ;

[0045] c33. And calculate the average deviation value based on multiple results, and then supplement the average deviation value into the prediction point M to obtain a new prediction point N.

[0046] Preferably, the formula for calculating the average deviation value in the c33 operation is:

[0047] Formula corresponding to Result 1: t 平 =(t1 + t2 + … + t x ) / x;

[0048] Formula corresponding to Result 2: t 平 =(t1 + t2 + … + t y ) / y;

[0049] Then the operation of introducing the average deviation value into the compensation of the prediction point M in the c24 operation to obtain a new prediction point N is:

[0050] Calculation corresponding to Result 1: T v = T k - t 平 ;

[0051] Calculation corresponding to Result 2: T v = T k + t 平 , and T v is the adjustment error time of the adjusted prediction point N.

[0052] Preferably, the operation steps for realizing coordinated control after the load command changes by introducing a real-time data set in the prediction compensation unit are:

[0053] D1. According to the trained coordinated analysis model, introduce the real-time data set and analyze it in combination with the load command change operation;

[0054] D2. That is, when the load command changes, calculate the corresponding deviation time, and then, in combination with the start time of the steam turbine main control response and the regenerative compensation situation, predict the actual prediction point based on the obtained deviation time;

[0055] D3. Then, based on the prediction point, realize the command control of the boiler main control to complete the automatic coordinated control of the regenerative control.

[0056] The present invention provides a coordinated control system based on disturbance rejection predictive control. Compared with the prior art, it has the following beneficial effects:

[0057] (1) The coordinated control system based on disturbance rejection predictive control, by setting up a coordinated control system, after completing the acquisition and processing of data, uses a coordinated analysis model to achieve the training compensation of historical data, and then introduces real-time data to achieve active coordination after the load command changes. Combining the error values reflected between compensation devices and the disturbance deviation values of the regenerative main control, it can find the predictive points that can be coordinated in advance, realizing synchronous coordinated control processing operations, not only effectively improving the coordination efficiency, but also ensuring the stability during the coordination process, effectively reducing the error impact, and improving the fault tolerance rate of coordinated control.

[0058] (2) The coordinated control system based on disturbance rejection predictive control, through the data analysis unit, uses an extraction function to process the initial data, extracts corresponding parameters from the initial data according to the required data requirements, forms the required historical data and real-time data, and directly completes the extraction of parameters from the initial data through subsequent demand data requirements and demand data conditions, ensuring the accuracy of data extraction while effectively removing the influence of useless data, thereby effectively improving the efficiency of data analysis.

[0059] (3) The coordinated control system based on disturbance rejection predictive control, by extracting the parameter data corresponding to different load commands as actual values, and then during the process of load command change, using the coordinated analysis model for analysis and simulation, analyzing the influence of the boiler main control coordinated control after the steam turbine main control responds, and analyzing the influence of the extraction amount of the regenerative main control on the coordinated control, so as to combine the influence of multi-objective factors to calculate and compensate for errors, thereby accurately and effectively obtaining the predictive points during real-time data coordination, and thus completing the advance coordinated processing and reducing errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is the principle block diagram of the coordinated control system of the present invention;

[0061] Figure 2 is the operation flow chart of the data analysis unit of the present invention;

[0062] Figure 3 is the analysis operation flow chart of the coordinated analysis model of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0064] Please refer to Figures 1 - 3, the present invention provides two technical solutions:

[0065] Embodiment 1. A coordinated control system based on disturbance rejection predictive control, comprising:

[0066] A data acquisition unit, configured to collect real-time operation data of a boiler and a steam turbine, and extract historical operation data;

[0067] A wireless transmission module, which uses wireless communication technology to implement data transmission and stores the collected data in a database;

[0068] A coordinated control system, which realizes data analysis and processing, and performs coordinated control according to a load command, specifically including:

[0069] A data analysis unit, which realizes data classification operations to form required historical data sets and real-time data sets;

[0070] A prediction compensation unit, which establishes a coordinated analysis model, uses the historical data set to train the coordinated analysis model, and then introduces the real-time data set to perform coordinated control operations after the load command changes, and marks the predicted control points;

[0071] An optimization transmission unit, which completes synchronous coordinated control operations and compensates for delay errors by implementing data transmission operations when reaching the control points;

[0072] A controllable display interface, which realizes control operations of various devices through the interface and displays data changes during operation.

[0073] By setting up a coordinated control system, after collecting and processing data, the coordinated analysis model is used to realize training compensation of historical data, and then real-time data is introduced to achieve active coordination after the load command changes. Combining the error values reflected between compensation devices and the disturbance deviation value of the regenerative main control, the predicted points that can be coordinated in advance are found, and synchronous coordinated control processing operations are realized, which not only effectively improves the coordination efficiency, but also ensures the stability during coordination, effectively reduces the error impact, and improves the fault tolerance rate of coordinated control.

[0074] In the embodiment of the present invention, the operation of data classification in the data analysis unit is as follows:

[0075] A1. First, convert the data signals collected by the sensor into digital signals that can be recognized and processed, and then summarize them into initial data;

[0076] A2. Process the initial data through an extraction function, extract corresponding parameters from the initial data based on the required data requirements, and form the required historical data and real-time data;

[0077] A3. Classify historical data and real-time data respectively to form the required historical data set labeled as P and real-time data set labeled as Q.

[0078] In the embodiment of the present invention, the processing operation of the extraction function in A2 is as follows:

[0079] a21. Set the extraction function F(u, v), where u is the requirement for parameters in the initial data to be extracted, and v is the condition for parameters in the initial data to be extracted;

[0080] a22. Use the subsequent required data category as the extraction requirement of the extraction function F(u, v), and the parameter names under the subsequent required data category as the extraction condition of the extraction function F(u, v);

[0081] a23. Then match the extraction requirement of the extraction function F(u, v) with the content in the initial data, and then match it with the extraction condition of the extraction function F(u, v) in the same initial data based on the match, so as to extract and classify the matched parameters to form the required data set.

[0082] In the embodiment of the present invention, the operation of establishing a coordinated analysis model in the prediction compensation unit is as follows:

[0083] B1. Extract the corresponding equipment data in the historical data and extract the operation connection relationship data between the equipment;

[0084] B2. Then integrate the data to form a coordinated analysis model for realizing the simulation control operations of the steam turbine main control, boiler main control, and regenerative heat main control, and then introduce the extracted historical data into the coordinated analysis model for training and analysis.

[0085] Among them, the data analysis unit processes the initial data through the extraction function, extracts the corresponding parameters based on the required data requirements from the initial data to form the required historical data and real-time data, and directly completes the extraction of parameters in the initial data through the subsequent demand data requirements and demand data conditions, ensuring the accuracy of data extraction while effectively removing the influence of useless data, thereby effectively improving the efficiency of data analysis.

[0086] In the embodiment of the present invention, the operation of the coordinated analysis model analyzing the historical data set is as follows:

[0087] C1. Extract the parameter data corresponding to different load commands as actual values, and then analyze and simulate using the coordinated analysis model during the change process of the load command;

[0088] C2. When a load change command is generated, realize the operation of the steam turbine main control, and analyze the influence of the coordinated control of the boiler main control after the steam turbine main control responds;

[0089] When a load change command is generated and after the operation of the steam turbine main control is completed, analyze the influence of the extraction amount of the regenerative main control on coordinated control.

[0090] In the embodiment of the present invention, the steps of analyzing the influence of the coordinated control of the boiler main control after the steam turbine main control responds in C2 are as follows:

[0091] c21. When the difference value of the load change command is J k At this time, the load change command forms a steam turbine load command and a boiler load command, and is transmitted to the steam turbine main control and the boiler main control to generate commands to control each sub-control system;

[0092] c22. When simulating the coordinated analysis model, the time data from the response of the steam turbine main control to the completion of the regulation operation is marked as T m Extract, and the time data from the coordinated response of the boiler main control to the completion of the regulation is T n Extract, and ensure the pressure balance between the front of the steam turbine and inside the boiler during the control process;

[0093] c23. Then, according to the time T m of the coordination of the steam turbine main control and the time T n of the coordination of the boiler main control, as well as the difference value J k of the load change command, determine the change coefficient of the deviation regulation, marked as L i ;

[0094] c24. Then, through multiple load change commands and repeating the operations of c21 to c23, obtain multiple change coefficients L i , and then obtain the average change coefficient corresponding to the load change command, and calculate the adjustment error time under the corresponding load change command based on the average change coefficient, so as to determine the prediction point when the boiler main control coordinates and responds to receive the command transmitted by the steam turbine main control, marked as M.

[0095] Among them, by extracting the parameter data corresponding to different load commands as actual values, and then during the process of the change of the load command, using the coordinated analysis model to analyze and simulate, analyze the influence of the coordinated control of the boiler main control after the steam turbine main control responds, analyze the influence of the extraction amount of the regenerative main control on coordinated control, so as to combine the influence of multi-objective factors to calculate and compensate for errors, so as to accurately and effectively obtain the prediction point during real-time data coordination, and thus complete the early coordinated processing and reduce errors.

[0096] In the embodiment of the present invention, the calculation formula of the change coefficient L i in c23 is:

[0097] L i =|T m -T n | / J k ;

[0098] i, m, n, and k represent the corresponding nth category parameters;

[0099] And the calculation formula for the average change coefficient in c24 is: L 平 =(L1 + L2 + … + L i ) / i;

[0100] Then, according to the average change coefficient, the difference J under the corresponding load change command is calculated, and k the adjustment error time of is: T k = L 平 × J k ;

[0101] At this time, the time point of the prediction point M is the self-regulation time T starting from the time when the steam turbine main control responds, k and the command is transmitted to achieve the coordinated response adjustment of the boiler main control.

[0102] By calculating the average change coefficient and combining multiple change coefficients to optimize the average value, not only can the error tolerance be improved, but also the adjustment error time after synthesizing multiple data is more accurate, and it is also convenient to apply to different scenarios for coordinated operation.

[0103] In the embodiment of the present invention, the steps for analyzing the influence of the extraction amount of the C3 regenerative main control on the coordinated control are as follows:

[0104] c31. According to the setting in c21 operation, when the difference of the load change command is J k , the coordinated control between the steam turbine main control and the boiler main control is carried out through the regenerative main control;

[0105] c32. Judge the parameter numerical result of the load command change;

[0106] Result 1: If the parameter value of the load command change becomes larger, the operation control of the regenerative main control shortens the time for coordinating between the boiler main control and the steam turbine main control, and the corresponding shortened time of the regenerative main control is t x ;

[0107] Result 2: If the parameter value of the load command change becomes smaller, the operation control of the regenerative main control increases the time for coordinating between the boiler main control and the steam turbine main control, and the corresponding increased time of the regenerative main control is t y ;

[0108] c33. And calculate the average deviation value according to multiple results, and then supplement the average deviation value into the prediction point M to obtain a new prediction point N.

[0109] The regenerative main control extracts hot steam from the steam turbine and then transfers it to the boiler for coordinated heating operations, thereby achieving faster and more efficient heating operations and effectively saving heating time.

[0110] In the embodiment of the present invention, the formula for calculating the average deviation value in the c33 operation is:

[0111] Formula corresponding to Result 1: t 平 =(t1 + t2 + … + t x ) / x;

[0112] Formula corresponding to Result 2: t 平 =(t1 + t2 + … + t y ) / y;

[0113] Then, the average deviation value is introduced into the compensation of the prediction point M in the c24 operation to obtain the new prediction point N operation:

[0114] Calculation corresponding to Result 1: T v =T k -t 平 ;

[0115] Calculation corresponding to Result 2: T v =T k +t 平 , and T v is the adjustment error time of the adjusted prediction point N.

[0116] The two average deviation values t 平 in the above are for data results in different situations, and for faster and more efficient heating operations, or for the control operation, there is an impact on the rate, so the adjustment error time of the prediction point N corresponding to different results is generated. For different instruction changes, different coordination methods are adopted.

[0117] In the embodiment of the present invention, the steps for implementing coordinated control operations after the load instruction changes by introducing a real-time data set in the prediction compensation unit are as follows:

[0118] D1. According to the trained coordinated analysis model, introduce the real-time data set and analyze it in combination with the load instruction change operation;

[0119] D2. That is, when the load instruction changes, calculate the corresponding deviation time, and then, in combination with the start time of the steam turbine main control response and the regenerative compensation situation, predict the actual prediction point based on the obtained deviation time;

[0120] D3. Then, based on the prediction point, implement the instruction control of the boiler main control to complete the automatic coordinated control of the regenerative control.

[0121] Embodiment 2. The difference compared with Embodiment 1 is that: the existing coordinated control system and the coordinated control system of the present invention are applied to multiple thermal power regenerative control operations, and the time required for coordinated control is recorded. Then, the actual load value after coordination is recorded and compared with the target load value. The specific comparison results are shown in Table 1:

[0122] Table 1 Comparison Results Table

[0123] Average time taken to complete coordination Accuracy rate of actual results Existing coordination control system 5 min 81％ Coordination control system of the present invention 2 min 99％

[0124] In summary, when the coordinated control system of the thermal power unit of the present invention is applied, the time required for coordination is greatly shortened. And after early prediction, the final result is more in line with the set target result value. Therefore, the application scope of the coordinated control system of the thermal power unit of the present invention in practical applications is wider and the effect is better.

[0125] At the same time, the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0126] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0127] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A coordinated control system based on disturbance suppression predictive control, characterized in that: include: Data acquisition unit, used to collect real-time operating data of boilers and steam turbines, and extract historical operating data; Wireless transmission module, which uses wireless communication technology to realize data transmission and stores the collected data in the database; Coordinated control system to realize data analysis and processing, and coordinate control according to load instructions, including: Data analysis unit, which implements data classification operations to form the required historical data sets and real-time data sets; The prediction and compensation unit establishes a coordination analysis model, uses historical data sets to train the coordination analysis model, and then introduces real-time data sets to implement coordinated control operations after load instruction changes, marking predicted control points; Optimize the transmission unit to complete the synchronous coordinated control operation and compensate for the delay error by realizing the data transmission operation when reaching the control point; The controllable display interface realizes the control operations of various devices through the interface, and displays the data changes during operation.

2. A coordinated control system based on disturbance suppression predictive control according to claim 1, characterized in that: The operations for data classification in the data analysis unit are: A1. First, the data signals collected by the sensor are converted into digital signals that can be recognized and processed, and then summarized into initial data; A2. Process the initial data through the extraction function, extract the corresponding parameters based on the initial data according to the required data requirements, and form the required historical data and real-time data; A3. Classify the historical data and real-time data into the required historical data set marked as P and the real-time data set marked as Q.

3. A coordinated control system based on disturbance suppression predictive control according to claim 2, characterized in that: The processing operation of the extraction function in A2 is: a21. Set the extraction function F(u, v), where u is the requirement for extracting the parameters in the initial data, and v is the condition for extracting the parameters in the initial data; a22. The data category required subsequently is used as the extraction requirement of the extraction function F(u, v), and the parameter name under the data category required subsequently is used as the extraction condition of the extraction function F(u, v); a23. Then, the extraction requirements of the extraction function F(u, v) are matched with the content in the initial data, and then the extraction conditions of the extraction function F(u, v) are matched based on the matching same initial data, so as to extract and classify the matched parameters to form the required data set.

4. The coordinated control system based on disturbance suppression predictive control according to claim 1, characterized in that: The operation of establishing the coordination analysis model in the prediction and compensation unit is: B1. Extract the corresponding equipment data from the historical data, and extract the operation connection relationship data between the equipment; B2. The data is then integrated to form a coordinated analysis model to realize the simulated control operations of the turbine master control, boiler master control and heat recovery master control. The extracted historical data is then introduced into the coordinated analysis model for training and analysis.

5. The coordinated control system based on disturbance suppression predictive control according to claim 1, characterized in that: The operation of the coordinated analysis model to analyze the historical data set is: C1. By extracting the parameter data corresponding to different load instructions as actual values, and then using the coordination analysis model to analyze and simulate according to the process of load instruction changes; C2. When the load change command is generated, the operation of the steam turbine master control is realized, and the influence of the boiler master control coordinated control after the steam turbine master control response is analyzed; C3. When the load change command is generated, after completing the operation of the turbine master control, analyze the impact of the extraction amount of the heat recovery master control on the coordinated control.

6. A coordinated control system based on disturbance suppression predictive control according to claim 5, characterized in that: The analysis in C2 shows that the steps of boiler master control coordination control after the turbine master control response are: c21, when the difference of load change command is J k When the load change instruction forms the turbine load instruction and the boiler load instruction, and is transmitted to the turbine master control and the boiler master control to generate instructions to control various sub-control systems; c22. The time from the main control response of the steam turbine to the completion of the control operation when the data is introduced into the coordinated analysis model simulation is marked as T m Extract, the time data from the boiler master control coordination response to the completion of regulation is T n Extraction, and during the control process, ensure the pressure balance between the turbine front and the boiler; c23, and then according to the time T coordinated by the steam turbine master control m Time T for coordination with boiler master control n The difference between the load change command and the load change command J k Determine the coefficient of variation of the deviation control and label it as L i ; c24, then through multiple load change instructions, and repeat the operations from c21 to c23, to obtain multiple change coefficients L i Then, the average change coefficient of the corresponding load change instruction is obtained, and the adjustment error time under the corresponding load change instruction is calculated based on the average change coefficient, so as to determine the prediction point marked as M when the boiler master control coordinates the response to receive the transmission instruction of the turbine master control.

7. A coordinated control system based on disturbance suppression predictive control according to claim 6, characterized in that: The coefficient of variation of c23 is L i The calculation formula is: L i =|T m -T n | / J k ; i, m, n and k represent the corresponding category parameters; And the corresponding average coefficient of variation in c24 is calculated as: L 平 =(L1+L2+…+L i ) / i; Then, the difference J corresponding to the load change instruction is calculated based on the average change coefficient. k The adjustment error time is: T k =L 平 ×J k ; At this time, the time point of the prediction point M is the time T from the start of the self-regulation of the turbine master control response. k Start transmitting instructions to achieve coordinated response and regulation of the boiler master control.

8. The coordinated control system based on disturbance suppression predictive control according to claim 7, characterized in that: The steps of C3 analyzing the influence of the extraction amount of the heat recovery main control on the coordinated control are: c31, according to the setting in c21 operation, when the difference of load change command is J k When the steam turbine master control and boiler master control are coordinated and controlled through the heat recovery master control; c32, judging the parameter value result of load instruction change; Result 1: If the parameter value of the load command change becomes larger, the operation control of the heat recovery master control to achieve the coordination time between the boiler master control and the turbine master control is reduced, and the corresponding reduction time of the heat recovery master control is t x ; Result 2: If the parameter value of the load command change becomes smaller, the time required for the heat recovery master control to achieve coordination between the boiler master control and the turbine master control will increase, and the corresponding increase in the heat recovery master control time is t y ; c33, and calculate the average deviation value based on multiple results, and then add the average deviation value to the prediction point M to obtain a new prediction point N.

9. A coordinated control system based on disturbance suppression predictive control according to claim 8, characterized in that: The formula for calculating the average deviation value in the c33 operation is: The formula corresponding to result 1 is: t 平 =(t1+t2+…+t x ) / x; The formula corresponding to result 2 is: t 平 =(t1+t2+…+t y ) / y; Then the average deviation value is introduced into the compensation of the prediction point M in the c24 operation to obtain the new prediction point N operation: The calculation corresponding to result 1 is: T v =T k -t 平 ; The calculation corresponding to result 2 is: T v =T k +t 平 , and T v is the adjustment error time of the adjusted prediction point N.

10. The coordinated control system based on disturbance suppression predictive control according to claim 1, characterized in that: The predictive compensation unit introduces a real-time data set to implement coordinated control after the load instruction changes. The operation steps are: D1. Based on the trained coordination analysis model, the real-time data set is introduced and analyzed in combination with the load instruction change operation; D2, that is, when the load command changes, the corresponding deviation time is converted, and then the actual prediction point is predicted based on the obtained deviation time in combination with the start time of the steam turbine master control response and the heat recovery compensation situation; D3. Then, based on the predicted point, the command control of the boiler master control is realized to complete the automatic coordinated control of the heat recovery control.

Images