Deep peak regulation dry-wet conversion control system based on extended state observer

Abstract

The invention relates to the technical field of large-scale thermal power generating unit control systems, in particular to a deep peak regulation dry-wet conversion control system based on an extended state observer, and aims to solve the problems that high-precision internal state and external disturbance estimation cannot be realized, state estimation cannot be updated in real time, and the control precision is low in the prior art. And dynamic response and stability cannot be ensured. An extended state vector is introduced through an extended state observer module, high-precision internal state and external disturbance estimation is realized, control precision and robustness are improved, parameters are adaptively adjusted in combination with a BP neural network, response speed and control performance are improved, state estimation is updated in real time, dynamic response and stability are ensured, and the method is suitable for large-scale popularization and application. The method is suitable for a complex nonlinear system, the control performance and the adaptive capacity are remarkably improved, and stable and reliable operation is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems for large thermal power units, and more specifically to a deep peak regulation dry-wet conversion control system based on an extended state observer. Background Art

[0002] During the operation of large-scale thermal power units, dry-wet conversion is an important operation process, and its control effect directly affects the stability and efficiency of the unit; traditional control systems have problems such as slow response speed, poor stability, and insufficient accuracy during deep peak regulation, and cannot meet the needs of modern thermal power units for efficient and stable operation.

[0003] The patent application with reference publication number CN115826651A discloses a deep peak-shaving dry-wet conversion control system based on an extended state observer, and the method includes the following steps: Step 1, collecting the water inlet flow, water tank water level, actual enthalpy value, water inlet pressure, and coal supply during the dry-wet conversion process of the thermal power unit; Step 2, building a water tank water level control system based on the extended state observer, using the extended state observer to estimate the water inlet flow and the water level of the water tank, and then controlling the flow opening through PID; Step 3, realizing automatic control of wet state to dry state and dry state to wet state through the dry-wet conversion control system; Step 4: introducing weight factors to perform online correction on the parameters of the extended state observer model during the load variation process; The control strategy effectively improves the control quality of the coordinated control system of large thermal power units under deep peak-shaving, and the control system has fast response speed, good stability performance, high accuracy, and good robustness and certain adaptability;

[0004] However, the above-mentioned reference patent adaptively adjusts the parameters of the expanded state observer through the BP neural network, which significantly improves the control performance and adaptability of large thermal power units under deep peak-shaving conditions, but cannot achieve high-precision internal state and external disturbance estimation, cannot update state estimation in real time, cannot guarantee dynamic response and stability, cannot accurately determine whether the dry and wet states need to be switched, reduces the flexibility and safety of the dry-wet state conversion, and cannot comprehensively evaluate the system operating status, cannot generate safety warnings in time and take corresponding protection measures.

[0005] To this end, we propose a deep peak-shaving dry-wet conversion control system based on an extended state observer to address the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a deep peak-shaving dry-wet conversion control system based on an extended state observer, which solves the problems that the prior art cannot achieve high-precision internal state and external disturbance estimation, cannot update state estimation in real time, cannot guarantee dynamic response and stability, cannot accurately determine whether the dry-wet state needs to be switched, reduces the flexibility and safety of the dry-wet state conversion, and cannot comprehensively evaluate the system operation status, cannot generate safety warnings in time and take corresponding protection measures.

[0007] The purpose of the present invention is achieved through the following technical solutions:

[0008] A deep peak load regulation dry-wet conversion control system based on an extended state observer includes a conversion control platform, a data acquisition module, an extended state observer module, a control law design module, a dry-wet conversion module, and a safety protection module;

[0009] The data acquisition module collects the system's operating data within the time threshold in real time through sensors, filters and calibrates the collected operating data, and feeds the processed operating data back to the conversion control platform;

[0010] The extended state observer module estimates the internal state variables of the system in real time and estimates the impact of external disturbances;

[0011] The control law design module selects the appropriate control algorithm according to the control objective and adjusts the controller parameters to optimize the system performance based on the state and disturbance estimation information provided by the extended state observer;

[0012] The dry-wet conversion module automatically or manually switches between dry and wet operation modes according to the system's operating data;

[0013] The safety protection module monitors the system status in real time, detects abnormal behavior and potential failures, and takes safety protection measures.

[0014] As a preferred embodiment of the present invention, the specific process of the extended state observer module estimating the internal state variables of the system and the influence of external disturbances is as follows:

[0015] Obtain the system's internal state variables in multiple sub-time periods and construct a state space model based on the internal state variables:

[0016]

[0017] where x is the output of the system, is the rate of change of the state variable x over time, x (n) is the nth derivative of x, u is the control input of the system, d is the unknown external disturbance, t is time, and f is a nonlinear function;

[0018] The unknown external disturbance d is regarded as the extended state of the system, denoted by x n+1 =d;

[0019] Assume that the derivative of the disturbance Bounded, that is, |h(t)|≤c, where c is a constant, so the state vector of the extended system is

[0020] The state equation of the extended system is:

[0021]

[0022] Design an extended state observer to estimate the state of the extended system

[0023] in is the state variable x n-1 The estimated value of is the estimated value of the disturbance d, x is the output of the system, that is, x = x1, β m (m=1, 2, ..., n+1) is the observer gain.

[0024] As a preferred embodiment of the present invention, the specific process of the control law design module selecting the control algorithm and adjusting the controller parameters is as follows:

[0025] Select the control algorithm based on the system characteristics and control objectives. The control algorithm selection strategy is as follows:

[0026] For simple systems, where steady-state accuracy is not a high requirement, the P control algorithm is selected.

[0027] For systems that need to eliminate steady-state errors, choose the PI control algorithm;

[0028] For systems that require fast response, overshoot suppression, and high precision, select the PID control algorithm;

[0029] For systems with model uncertainty and external disturbances, choose a robust control algorithm;

[0030] For nonlinear systems, it is difficult to establish an accurate mathematical model for the system, so fuzzy control algorithm is selected;

[0031] For systems that require optimal performance indicators, select the optimal control algorithm.

[0032] As a preferred embodiment of the present invention, the temperature estimation values ​​inside the system in multiple sub-time periods are obtained, and the arithmetic mean of the multiple temperature estimation values ​​obtained is calculated, and the arithmetic mean of the multiple temperature estimation values ​​is recorded as the temperature deviation value WPZ;

[0033] Obtain the pressure estimation values ​​inside the system in multiple sub-time periods, thereby constructing a set A of pressure estimation values, and record the average of the difference between the maximum subset and the minimum subset in set A as the pressure deviation value YPZ;

[0034] Obtain the estimated flow values ​​within the system in multiple sub-time periods, and use them to construct a set B of flow deviation values, obtain the mean in set B, and mark the mean in set B as the flow deviation mean. Establish a rectangular coordinate system with time as the X-axis and the flow deviation mean as the Y-axis, draw a flow deviation mean curve by plotting points, and draw a preset flow deviation mean threshold curve in the coordinate system at the same time. Obtain the product value obtained by normalizing the time length corresponding to the line segment of the flow deviation mean curve above the preset flow deviation mean threshold curve and the area enclosed by the upper line segment and the preset flow deviation mean threshold curve from the coordinate system, and mark it as the flow anomaly risk value LYZ;

[0035] The estimated values ​​of the disturbance d in multiple sub-time periods are obtained, and the arithmetic mean of the obtained multiple estimated values ​​of the disturbance d is calculated, and the absolute value of the arithmetic mean of the multiple estimated values ​​of the disturbance d is recorded as the average disturbance deviation value PRP.

[0036] As a preferred embodiment of the present invention, the temperature deviation value WPZ, the pressure deviation value YPZ, the flow abnormality risk value LYZ and the average disturbance deviation value PRP are obtained, and the adjustment evaluation coefficient TPX is calculated by the following formula:

[0037]

[0038] Among them, e1, e2, e3 and e4 are all preset proportional factor coefficients, e4>e3>e2>e1>0, and the adjustment evaluation coefficient TPX is compared and analyzed with the preset adjustment evaluation coefficient threshold:

[0039] If the adjustment evaluation coefficient TPX is greater than or equal to the preset adjustment evaluation coefficient threshold, it indicates that the system performance index does not reach the expected performance index, and the controller parameters need to be adjusted to optimize the system performance;

[0040] If the adjustment evaluation coefficient TPX is less than the preset adjustment evaluation coefficient threshold, it indicates that the system performance index has reached the expected performance index and there is no need to adjust the controller parameters to optimize the system performance.

[0041] As a preferred embodiment of the present invention, the specific process of the dry-wet conversion module automatically or manually switching between the dry and wet operation modes is as follows:

[0042] Obtain the operating condition data after filtering and calibration;

[0043] The real-time temperature of the system is obtained, and the difference between the real-time temperature and the median of the corresponding preset suitable temperature range is calculated and the absolute value is taken to obtain the real-time temperature difference, and the real-time temperature difference is marked as WDC;

[0044] The real-time pressure of the system is obtained, and the difference between the real-time pressure and the median of the corresponding preset suitable pressure range is calculated and the absolute value is taken to obtain the real-time pressure difference, and the real-time pressure difference is marked as YLC;

[0045] The real-time flow of the system is obtained, and the difference between the real-time flow and the median of the corresponding preset suitable flow range is calculated and the absolute value is taken to obtain the real-time flow difference, and the real-time flow difference is marked as LLC;

[0046] The state transition assessment coefficient ZZP is calculated by the following formula:

[0047] ZZP = k1*WDC+k2*YLC+k3*LLC;

[0048] Among them, k1, k2 and k3 are all weight coefficients, k1+k2+k3=1, and the state transition assessment coefficient ZZP is compared and analyzed with the preset state transition assessment coefficient threshold:

[0049] If ZZP is greater than the preset state transition assessment coefficient threshold, it indicates that the current mode needs to be switched;

[0050] If ZZP is less than or equal to the preset state transition assessment coefficient threshold, it indicates that the current mode does not need to be switched.

[0051] As a preferred embodiment of the present invention, according to the mode switching judgment result, the actuator is controlled to switch between the dry state and the wet state mode, and the mode operation logic is as follows:

[0052] Dry mode: close the valves related to the wet side and open the valves related to the dry side;

[0053] Wet mode: close the valves related to the dry side and open the valves related to the wet side;

[0054] A manual switch is set to allow the operator to manually switch between dry and wet operation modes according to actual conditions. Manual switching has a higher priority than automatic switching.

[0055] According to the automatic switching logic and manual switching instructions, the switching action of the dry-wet conversion module is controlled.

[0056] As a preferred implementation of the present invention, the specific process of the security protection module monitoring the system status in real time and taking security protection measures is as follows:

[0057] Get the actual values ​​of the system state variables within the time threshold to construct the feature matrix J = [W, Y, L]T , where W is the actual value of temperature, Y is the actual value of pressure, and L is the actual value of flow;

[0058] Obtain the estimated values ​​of the system state variables within the time threshold to construct the feature matrix in is the estimated value of temperature, is the estimated value of pressure, is the estimated value of flow rate;

[0059] The root mean square value RMSE of the state variable estimation error is calculated by the following formula:

[0060]

[0061] The switching duration when the dry-wet switching is performed within the time threshold is obtained, and the arithmetic average of the obtained multiple switching durations is calculated, and the arithmetic average of the multiple switching durations is recorded as the average switching duration value QHS.

[0062] As a preferred embodiment of the present invention, the control signal v(t) of the controller within the time threshold is obtained to generate a control period T K , the control signal is obtained in the control period T through the following formula K The signal change rate within:

[0063]

[0064] Calculate the arithmetic mean of the calculated multiple signal change rates, and record the arithmetic mean of the multiple signal change rates as the average signal change rate XBL;

[0065] If RMSE is less than the preset RMS value threshold of the state variable estimation error, QHS is less than the preset average switching duration value threshold, and XBL is less than the preset average signal change rate threshold, it indicates that the system is in normal operation and no signal is generated;

[0066] If RMSE is greater than or equal to the preset root mean square value threshold of the state variable estimation error, or QHS is greater than or equal to the preset average switching duration value threshold, or XBL is greater than or equal to the preset average signal change rate threshold, it indicates that there are abnormal behaviors and potential faults in the system operation, and a safety warning signal is generated and sent to the conversion control platform.

[0067] As a preferred implementation of the present invention, after receiving the safety warning signal, the conversion control platform immediately takes corresponding safety protection measures, and the safety protection measures are as follows:

[0068] Immediately stop the dry-wet conversion process;

[0069] Isolate the faulty equipment or subsystem from the system;

[0070] For key equipment, configure a backup system to automatically switch to the backup system when the main system fails;

[0071] Allows operators to manually intervene in emergency situations;

[0072] Start the troubleshooting program.

[0073] Compared with the prior art, the advantages of the present invention are:

[0074] (1) In the present invention, an extended state vector is introduced by an extended state observer module to achieve high-precision estimation of internal state and external disturbance, improve control accuracy and robustness, and combine with BP neural network to adaptively adjust parameters, improve response speed and control performance, update state estimation in real time, ensure dynamic response and stability, and be applicable to complex nonlinear systems, significantly improve control performance and adaptive ability, and ensure stable and reliable operation;

[0075] (2) In the present invention, the temperature, pressure and flow rate differences are calculated in real time through the dry-wet conversion module, and the mode is automatically switched according to the comparison between the state conversion evaluation coefficient and the preset threshold value; at the same time, a manual switching switch is provided to improve flexibility and safety, and manual operation is given priority. Multi-dimensional data processing and dynamic mode adjustment ensure efficient and stable operation of the system, reduce misjudgment, and improve reliability;

[0076] (3) In the present invention, the system status is monitored in real time through the safety protection module, the root mean square value of the state variable estimation error, the average switching time and the average signal change rate are calculated, and a comprehensive evaluation is made to generate a safety warning; after receiving the warning, the platform immediately takes protection measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0078] Figure 2 This is a system block diagram of Embodiment 2 of the present invention;

[0079] Figure 3 This is a logical flow diagram of the first embodiment of the present invention;

[0080] Figure 4 This is a logical flow diagram of the second embodiment of the present invention. DETAILED DESCRIPTION

[0081] The following will combine the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all the embodiments. All other embodiments obtained by ordinary technicians in this field without creative work based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0082] Embodiment 1: Figure 1 and Figure 3 As shown, the deep peak load regulation dry-wet conversion control system based on the extended state observer proposed in the present invention includes a conversion control platform, a data acquisition module, an extended state observer module, a control law design module, a dry-wet conversion module and a safety protection module;

[0083] The conversion control platform is connected to the control law design module and the safety protection module in a two-way communication manner. The conversion control platform is connected to the dry-wet conversion module, the data acquisition module and the expanded state observer module in a one-way communication manner. The expanded state observer module is connected to the control law design module in a one-way communication manner. The control law design module is connected to the dry-wet conversion module in a one-way communication manner.

[0084] The data acquisition module collects the system's operating data within a time threshold in real time through sensors (the time threshold is the length of time the system runs for a period of time, and the time threshold is divided into g sub-time periods, where g is a natural number greater than zero). The operating data includes temperature, pressure, and flow. The collected operating data is filtered (using the Kalman filter method for filtering) and calibrated (using the polynomial fitting method for calibration), and the processed operating data is fed back to the conversion control platform;

[0085] By subdividing the time threshold into multiple sub-time periods, the system operating condition changes can be captured more accurately, ensuring high resolution and accuracy of the data. Kalman filtering is used to remove noise and improve the data signal-to-noise ratio. Polynomial fitting corrects sensor errors, enhances data consistency, and performs real-time data collection and processing to ensure rapid response of the control platform. This design improves data reliability and stability, reduces the risk of misjudgment and control failure, adapts to a variety of sensors and working conditions, and ensures system robustness and adaptability. Finally, the data acquisition module provides high-quality data to help the conversion control platform make effective decisions and improve system control performance and operational stability.

[0086] The extended state observer module estimates the system's internal state variables in real time, including temperature, pressure, and flow, and estimates the impact of external disturbances;

[0087] The specific process of the extended state observer module estimating the internal state variables of the system and the impact of external disturbances is as follows:

[0088] Obtain the system's internal state variables in multiple sub-time periods. The internal state variables include temperature, pressure, and flow. Construct a state space model based on the internal state variables:

[0089]

[0090] where x is the output of the system, such as temperature, pressure, and flow, is the rate of change of the state variable x over time, x (n) is the nth-order derivative of x, u is the control input of the system, d is the unknown external disturbance, t is time, and f is a nonlinear function used to describe the dynamic characteristics of the system;

[0091] The unknown external disturbance d is regarded as the extended state of the system, denoted by x n+1 =d;

[0092] Assume that the derivative of the disturbance Bounded, that is, |h(t)|≤c, where c is a constant, so the state vector of the extended system is

[0093] The state equation of the extended system is:

[0094]

[0095] Design an extended state observer to estimate the state of the extended system

[0096] in is the state variable x n-1 The estimated value of is the estimated value of the disturbance d, x is the output of the system, that is, x = x1, β m (m=1,2,…,n+1) is the observer gain, which needs to be adjusted according to the system characteristics;

[0097] By introducing the extended state vector, the extended state observer module can achieve high-precision estimation of the system's internal state variables and external disturbances, thereby improving control accuracy. The module has strong robustness to unknown disturbances and can effectively suppress the influence of external interference. Combined with the BP neural network, the observer's gain parameters can be adaptively adjusted to further improve the system's response speed and control performance. In addition, the module can continuously update the state estimate during real-time operation to ensure the dynamic response and stability of the system. It is suitable for various complex nonlinear systems, especially the control requirements of large thermal power units under deep peak-shaving conditions, which significantly improves the system's control performance and adaptability, and ensures the stable and reliable operation of thermal power units under deep peak-shaving conditions.

[0098] The control law design module selects the appropriate control algorithm according to the control objective and adjusts the controller parameters to optimize the system performance based on the state and disturbance estimation information provided by the extended state observer;

[0099] The specific process of the control law design module selecting the control algorithm and adjusting the controller parameters is as follows:

[0100] Select the control algorithm based on the system characteristics and control objectives. The control algorithm selection strategy is as follows:

[0101] For simple systems, where steady-state accuracy is not a high requirement, the P control algorithm is selected.

[0102] For systems that need to eliminate steady-state errors, choose the PI control algorithm;

[0103] For systems that require fast response, overshoot suppression, and high precision, select the PID control algorithm;

[0104] For systems with model uncertainty and external disturbances, choose a robust control algorithm;

[0105] For nonlinear systems, it is difficult to establish an accurate mathematical model for the system, so fuzzy control algorithm is selected;

[0106] For systems that require optimal performance indicators, select the optimal control algorithm;

[0107] Obtaining temperature estimation values ​​inside the system in multiple sub-time periods, calculating an arithmetic mean of the obtained multiple temperature estimation values, and recording the arithmetic mean of the multiple temperature estimation values ​​as a temperature deviation value WPZ;

[0108] Obtain the pressure estimation values ​​inside the system in multiple sub-time periods, thereby constructing a set A of pressure estimation values, and record the average of the difference between the maximum subset and the minimum subset in set A as the pressure deviation value YPZ;

[0109] Obtain the estimated flow values ​​within the system in multiple sub-time periods, and use them to construct a set B of flow deviation values, obtain the mean in set B, and mark the mean in set B as the flow deviation mean. Establish a rectangular coordinate system with time as the X-axis and the flow deviation mean as the Y-axis, draw a flow deviation mean curve by plotting points, and draw a preset flow deviation mean threshold curve in the coordinate system at the same time. Obtain the product value obtained by normalizing the time length corresponding to the line segment of the flow deviation mean curve above the preset flow deviation mean threshold curve and the area enclosed by the upper line segment and the preset flow deviation mean threshold curve from the coordinate system, and mark it as the flow anomaly risk value LYZ;

[0110] Obtain estimated values ​​of disturbance d in multiple sub-time periods, calculate the arithmetic mean of the obtained multiple estimated values ​​of disturbance d, and take the absolute value of the arithmetic mean of the multiple estimated values ​​of disturbance d as the average disturbance deviation value PRP;

[0111] The temperature deviation value WPZ, pressure deviation value YPZ, flow abnormality risk value LYZ and average disturbance deviation value PRP are obtained, and the adjustment assessment coefficient TPX is calculated by the following formula:

[0112]

[0113] Among them, e1, e2, e3 and e4 are all preset proportional factor coefficients, e4>e3>e2>e1>0, and the adjustment evaluation coefficient TPX is compared and analyzed with the preset adjustment evaluation coefficient threshold:

[0114] If the adjustment evaluation coefficient TPX is greater than or equal to the preset adjustment evaluation coefficient threshold, it indicates that the system performance index does not reach the expected performance index, and the controller parameters need to be adjusted to optimize the system performance;

[0115] If the adjustment evaluation coefficient TPX is less than the preset adjustment evaluation coefficient threshold, it indicates that the system performance index has reached the expected performance index and there is no need to adjust the controller parameters to optimize the system performance;

[0116] The control law design module can select appropriate control algorithms according to system characteristics and control objectives to ensure the accuracy and effectiveness of the control strategy; dynamically adjust controller parameters and optimize system performance by acquiring and processing state estimates in real time; comprehensively evaluate multiple factors such as temperature, pressure, flow and disturbance to ensure the comprehensiveness and accuracy of control; automatically adjust the control strategy according to the real-time status of the system and external disturbances to improve the system's adaptability and robustness; quickly determine whether the system performance meets the standards by adjusting the comparison between the evaluation coefficient TPX and the preset threshold to achieve efficient optimization; reduce the risk of misjudgment and control failure through multi-dimensional data processing and evaluation, and improve the reliability and stability of the system.

[0117] Embodiment 2: The technical solution of the embodiment of the present invention is different from that of embodiment 1 in that:

[0118] like Figure 2 and Figure 4 As shown, the dry-wet conversion module automatically or manually switches the dry and wet operation modes according to the system's operating data;

[0119] The specific process of the dry-wet conversion module automatically or manually switching between dry and wet operation modes is as follows:

[0120] Obtain the operating condition data after filtering and calibration;

[0121] The real-time temperature of the system is obtained, and the difference between the real-time temperature and the median of the corresponding preset suitable temperature range is calculated and the absolute value is taken to obtain the real-time temperature difference, and the real-time temperature difference is marked as WDC;

[0122] The real-time pressure of the system is obtained, and the difference between the real-time pressure and the median of the corresponding preset suitable pressure range is calculated and the absolute value is taken to obtain the real-time pressure difference, and the real-time pressure difference is marked as YLC;

[0123] The real-time flow of the system is obtained, and the difference between the real-time flow and the median of the corresponding preset suitable flow range is calculated and the absolute value is taken to obtain the real-time flow difference, and the real-time flow difference is marked as LLC;

[0124] The state transition assessment coefficient ZZP is calculated by the following formula:

[0125] ZZP = k1*WDC+k2*YLC+k3*LLC;

[0126] Among them, k1, k2 and k3 are all weight coefficients, k1+k2+k3=1. By analyzing historical data, we can understand the influence of weight coefficients k1, k2 and k3 on the dry-wet state switching. According to historical data, we can determine the specific values ​​of k1, k2 and k3 to ensure that the calculated state conversion evaluation coefficient ZZP can accurately determine whether the current dry-wet mode needs to be switched;

[0127] Compare and analyze the state transition assessment coefficient ZZP with the preset state transition assessment coefficient threshold:

[0128] If ZZP is greater than the preset state transition assessment coefficient threshold, it indicates that the current mode needs to be switched;

[0129] If ZZP is less than or equal to the preset state transition assessment coefficient threshold, it indicates that the current mode does not need to be switched;

[0130] According to the mode switching judgment result, the actuator (such as valve, switch, etc.) is controlled to switch between dry mode and wet mode. The mode operation logic is as follows:

[0131] Dry mode: close the valves related to the wet side and open the valves related to the dry side;

[0132] Wet mode: close the valves related to the dry side and open the valves related to the wet side;

[0133] A manual switch is set to allow the operator to manually switch between dry and wet operation modes according to actual conditions. Manual switching has a higher priority than automatic switching.

[0134] Control the switching action of the dry-wet conversion module according to the automatic switching logic and manual switching instructions;

[0135] The dry-wet conversion module comprehensively evaluates the system status by calculating the difference between real-time temperature, pressure and flow, and ensures the accuracy and effectiveness of mode switching. The dry-wet conversion module automatically determines whether the mode needs to be switched based on the comparison between the state conversion assessment coefficient ZZP and the preset threshold, thereby realizing intelligent control. The dry-wet conversion module is equipped with a manual switching switch, allowing the operator to manually switch the mode according to actual conditions, thereby improving the flexibility and adaptability of the system. The dry-wet conversion module reduces the risk of misjudgment and control failure and improves the reliability and stability of the system through multi-dimensional data processing and evaluation. The dry-wet conversion module acquires and processes operating condition data in real time, and dynamically adjusts the mode to ensure that the system can operate efficiently and stably under different operating conditions. The manual switching priority of the dry-wet conversion module is higher than the automatic switching, and the operator can easily intervene in the system operation to ensure the safety and controllability of the system.

[0136] Security protection module, which monitors system status in real time, detects abnormal behavior and potential failures, and takes security protection measures;

[0137] The specific process of the security protection module monitoring the system status in real time and taking security protection measures is as follows:

[0138] Get the actual values ​​of the system state variables within the time threshold to construct the feature matrix J = [W, Y, L] T , where W is the actual value of temperature, Y is the actual value of pressure, and L is the actual value of flow;

[0139] Obtain the estimated values ​​of the system state variables within the time threshold to construct the feature matrix in is the estimated value of temperature, is the estimated value of pressure, is the estimated value of flow rate;

[0140] The root mean square value RMSE of the state variable estimation error is calculated by the following formula:

[0141]

[0142] The switching time when the dry-wet switching is performed within the time threshold is obtained, and the arithmetic average of the multiple switching time lengths obtained is calculated, and the arithmetic average of the multiple switching time lengths is recorded as the average switching time length value QHS;

[0143] Get the control signal v(t) of the controller within the time threshold and generate the control period T K , the control signal is obtained in the control period T through the following formula K The signal change rate within:

[0144]

[0145] Calculate the arithmetic mean of the calculated multiple signal change rates, and record the arithmetic mean of the multiple signal change rates as the average signal change rate XBL;

[0146] If RMSE is less than the preset RMS value threshold of the state variable estimation error, QHS is less than the preset average switching duration value threshold, and XBL is less than the preset average signal change rate threshold, it indicates that the system is in normal operation and no signal is generated;

[0147] If RMSE is greater than or equal to the preset root mean square value threshold of the state variable estimation error, or QHS is greater than or equal to the preset average switching duration value threshold, or XBL is greater than or equal to the preset average signal change rate threshold, it indicates that the system operation has abnormal behavior and potential failures, and a safety warning signal is generated and sent to the conversion control platform;

[0148] After receiving the safety warning signal, the conversion control platform immediately takes corresponding safety protection measures. The safety protection measures are as follows:

[0149] Immediately stop the dry-wet conversion process and take emergency measures to protect the system safety;

[0150] Isolate the faulty equipment or subsystem from the system;

[0151] For key equipment, configure a backup system to automatically switch to the backup system when the main system fails;

[0152] Allows operators to manually intervene in emergency situations;

[0153] Start the fault diagnosis program, analyze the cause of the fault, and take corresponding corrective measures;

[0154] The safety protection module monitors the system status in real time to ensure timely detection of abnormal behaviors and potential faults. It comprehensively evaluates the system status by calculating the root mean square value of the state variable estimation error, the average switching duration value and the average signal change rate to ensure the accuracy of the evaluation. It generates a safety warning signal based on the evaluation results to realize intelligent abnormality detection and warning. After receiving the safety warning signal, the conversion control platform immediately takes corresponding safety protection measures to ensure the safe operation of the system. It provides multi-level safety protection through various measures such as stopping dry-wet conversion, isolating faulty equipment, switching to backup systems, manual intervention and fault diagnosis, allowing operators to perform manual intervention in emergency situations to ensure the safety and controllability of the system.

[0155] The working principle of the present invention is as follows: when in use, an extended state vector is introduced through an extended state observer module to achieve high-precision estimation of internal states and external disturbances, improve control accuracy and robustness, and combine with BP neural network to adaptively adjust parameters, improve response speed and control performance, update state estimation in real time, ensure dynamic response and stability, and be suitable for complex nonlinear systems. It significantly improves control performance and adaptive ability, ensures stable and reliable operation, and calculates temperature, pressure and flow differences in real time through a dry-wet conversion module, and automatically switches modes based on the comparison between a state conversion evaluation coefficient and a preset threshold. At the same time, a manual switching switch is provided to improve flexibility and safety, and manual operations are responded to first. Multi-dimensional data processing and dynamic mode adjustment ensure efficient and stable operation of the system, reduce misjudgment, and improve reliability. The system status is monitored in real time through a safety protection module, the root mean square value of the state variable estimation error, the average switching time and the average signal change rate are calculated, and a safety warning is comprehensively evaluated and generated. After the platform receives the warning, protective measures are taken immediately.

[0156] The above are only preferred specific implementation modes of the present invention; however, the protection scope of the present invention is not limited thereto; any technician familiar with the technical field within the technical scope disclosed by the present invention; any equivalent replacement or change based on the technical solution and improved concept of the present invention shall be covered within the protection scope of the present invention.

Claims

1. Deep peak load regulation dry-wet conversion control system based on extended state observer, characterized in that: It includes conversion control platform, data acquisition module, extended state observer module, control law design module, dry-wet conversion module and safety protection module; The data acquisition module collects the system's operating data within the time threshold in real time through sensors, filters and calibrates the collected operating data, and feeds the processed operating data back to the conversion control platform; The extended state observer module estimates the internal state variables of the system in real time and estimates the impact of external disturbances; The control law design module selects the appropriate control algorithm according to the control objective and adjusts the controller parameters to optimize the system performance based on the state and disturbance estimation information provided by the extended state observer; The dry-wet conversion module automatically or manually switches between dry and wet operation modes according to the system's operating data; The safety protection module monitors the system status in real time, detects abnormal behavior and potential failures, and takes safety protection measures.

2. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 1 is characterized in that: The specific process of the extended state observer module estimating the internal state variables of the system and the influence of external disturbances is as follows: Obtain the system's internal state variables in multiple sub-time periods and construct a state space model based on the internal state variables: where x is the output of the system, is the rate of change of the state variable x over time, x (n) is the nth-order derivative of x, u is the control input of the system, d is the unknown external disturbance, t is time, and f is a nonlinear function; The unknown external disturbance d is regarded as the extended state of the system, denoted by x n+1 =d; Assume that the derivative of the disturbance Bounded, that is, |h(t)|≤c, where c is a constant, so the state vector of the extended system is The state equation of the extended system is: Design an extended state observer to estimate the state of the extended system in is the state variable x n-1 The estimated value of is the estimated value of the disturbance d, x is the output of the system, that is, x = x1, β m (m=1, 2, ..., n+1) is the observer gain.

3. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 1 is characterized in that: The specific process of the control law design module selecting the control algorithm and adjusting the controller parameters is as follows: Select the control algorithm based on the system characteristics and control objectives. The control algorithm selection strategy is as follows: For simple systems, where steady-state accuracy is not a high requirement, the P control algorithm is selected. For systems that need to eliminate steady-state errors, choose the PI control algorithm; For systems that require fast response, overshoot suppression, and high precision, select the PID control algorithm; For systems with model uncertainty and external disturbances, choose a robust control algorithm; For nonlinear systems, it is difficult to establish an accurate mathematical model for the system, so fuzzy control algorithm is selected; For systems that require optimal performance indicators, select the optimal control algorithm.

4. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 2 is characterized in that: Obtaining temperature estimation values ​​inside the system in multiple sub-time periods, calculating an arithmetic mean of the obtained multiple temperature estimation values, and recording the arithmetic mean of the multiple temperature estimation values ​​as a temperature deviation value WPZ; Obtain the pressure estimation values ​​inside the system in multiple sub-time periods, thereby constructing a set A of pressure estimation values, and record the average of the difference between the maximum subset and the minimum subset in set A as the pressure deviation value YPZ; Obtain the estimated flow values ​​within the system in multiple sub-time periods, and use them to construct a set B of flow deviation values, obtain the mean in set B, and mark the mean in set B as the flow deviation mean. Establish a rectangular coordinate system with time as the X-axis and the flow deviation mean as the Y-axis, draw a flow deviation mean curve by plotting points, and draw a preset flow deviation mean threshold curve in the coordinate system at the same time. Obtain the product value obtained by normalizing the time length corresponding to the line segment of the flow deviation mean curve above the preset flow deviation mean threshold curve and the area enclosed by the upper line segment and the preset flow deviation mean threshold curve from the coordinate system, and mark it as the flow anomaly risk value LYZ; The estimated values ​​of the disturbance d in multiple sub-time periods are obtained, and the arithmetic mean of the obtained multiple estimated values ​​of the disturbance d is calculated, and the absolute value of the arithmetic mean of the multiple estimated values ​​of the disturbance d is recorded as the average disturbance deviation value PRP.

5. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 4 is characterized in that: The temperature deviation value WPZ, pressure deviation value YPZ, flow abnormality risk value LYZ and average disturbance deviation value PRP are obtained, and the adjustment assessment coefficient TPX is calculated by the following formula: Among them, e1, e2, e3 and e4 are all preset proportional factor coefficients, e4>e3>e2>e1>0, and the adjustment evaluation coefficient TPX is compared and analyzed with the preset adjustment evaluation coefficient threshold: If the adjustment evaluation coefficient TPX is greater than or equal to the preset adjustment evaluation coefficient threshold, it indicates that the system performance index does not reach the expected performance index, and the controller parameters need to be adjusted to optimize the system performance; If the adjustment evaluation coefficient TPX is less than the preset adjustment evaluation coefficient threshold, it indicates that the system performance index has reached the expected performance index and there is no need to adjust the controller parameters to optimize the system performance.

6. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 1 is characterized in that: The specific process of the dry-wet conversion module automatically or manually switching between the dry and wet operating modes is as follows: Obtain the operating condition data after filtering and calibration; The real-time temperature of the system is obtained, and the difference between the real-time temperature and the median of the corresponding preset suitable temperature range is calculated and the absolute value is taken to obtain the real-time temperature difference, and the real-time temperature difference is marked as WDC; The real-time pressure of the system is obtained, and the difference between the real-time pressure and the median of the corresponding preset suitable pressure range is calculated and the absolute value is taken to obtain the real-time pressure difference, and the real-time pressure difference is marked as YLC; The real-time flow of the system is obtained, and the difference between the real-time flow and the median of the corresponding preset suitable flow range is calculated and the absolute value is taken to obtain the real-time flow difference, and the real-time flow difference is marked as LLC; The state transition assessment coefficient ZZP is calculated by the following formula: ZZP = k1*WDC+k2*YLC+k3*LLC; Among them, k1, k2 and k3 are all weight coefficients, k1+k2+k3=1, and the state transition assessment coefficient ZZP is compared and analyzed with the preset state transition assessment coefficient threshold: If ZZP is greater than the preset state transition assessment coefficient threshold, it indicates that the current mode needs to be switched; If ZZP is less than or equal to the preset state transition assessment coefficient threshold, it indicates that the current mode does not need to be switched.

7. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 6 is characterized in that: According to the mode switching judgment result, the actuator is controlled to switch between dry and wet modes. The mode operation logic is as follows: Dry mode: close the valves related to the wet side and open the valves related to the dry side; Wet mode: close the valves related to the dry side and open the valves related to the wet side; A manual switch is set to allow the operator to manually switch between dry and wet operation modes according to actual conditions. Manual switching has a higher priority than automatic switching. According to the automatic switching logic and manual switching instructions, the switching action of the dry-wet conversion module is controlled.

8. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 2 is characterized in that: The specific process of the security protection module monitoring the system status in real time and taking security protection measures is as follows: Get the actual values ​​of the system state variables within the time threshold to construct the feature matrix J = [W, Y, L] T , where W is the actual value of temperature, Y is the actual value of pressure, and L is the actual value of flow; Obtain the estimated values ​​of the system state variables within the time threshold to construct the feature matrix in is the estimated value of temperature, is the estimated value of pressure, is the estimated value of the flow rate; The root mean square value RMSE of the state variable estimation error is calculated by the following formula: The switching duration when the dry-wet switching is performed within the time threshold is obtained, and the arithmetic average of the obtained multiple switching durations is calculated, and the arithmetic average of the multiple switching durations is recorded as the average switching duration value QHS.

9. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 8, characterized in that: Get the control signal v(t) of the controller within the time threshold and generate the control period T K , the control signal is obtained in the control period T through the following formula K The signal change rate within: Calculate the arithmetic mean of the calculated multiple signal change rates, and record the arithmetic mean of the multiple signal change rates as the average signal change rate XBL; If RMSE is less than the preset RMS value threshold of the state variable estimation error, QHS is less than the preset average switching duration value threshold, and XBL is less than the preset average signal change rate threshold, it indicates that the system is in normal operation and no signal is generated; If RMSE is greater than or equal to the preset root mean square value threshold of the state variable estimation error, or QHS is greater than or equal to the preset average switching duration value threshold, or XBL is greater than or equal to the preset average signal change rate threshold, it indicates that there are abnormal behaviors and potential faults in the system operation, and a safety warning signal is generated and sent to the conversion control platform.

10. The deep peak load regulation dry-wet conversion control system based on extended state observer according to claim 9, characterized in that: After receiving the safety warning signal, the conversion control platform immediately takes corresponding safety protection measures. The safety protection measures are as follows: Immediately stop the dry-wet conversion process; Isolate the faulty equipment or subsystem from the system; For key equipment, configure a backup system to automatically switch to the backup system when the main system fails; Allows operators to manually intervene in emergency situations; Start the troubleshooting program.

Images