Cooperative control method for intelligent turbine controller

Abstract

The invention relates to the technical field of intelligent turbine control, and discloses a cooperative control method for an intelligent turbine controller, and the method comprises the steps: introducing a time sequence analysis algorithm, analyzing the historical data of a control system, constructing a time delay prediction model, predicting the time delay characteristics of the system in real time, and calculating a feedforward compensation amount and a feedback control amount; by dynamically adjusting the time delay compensation amount and combining a multi-machine cooperative control strategy, the influence of time delay difference on the system performance is effectively reduced, and cooperative work among multiple intelligent turbine controllers is optimized; the method not only can respond to the change of the system in real time, but also can ensure that the controllers cooperatively work under the global optimal strategy, improves the stability and precision of the system, and is suitable for a complex working condition environment with a high dynamic response requirement.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent turbine control, and in particular to a collaborative control method for an intelligent turbine controller. Background Art

[0002] With the rapid development of industrial automation and intelligent technology, intelligent turbine controllers (ITCs), as important control system components, are widely used in turbine equipment in the energy, chemical, and manufacturing fields. The main function of intelligent turbine controllers is to adjust the operating status of turbine equipment according to the input control signal to ensure that the equipment operates efficiently at the optimal operating point; however, with the increase in system complexity and changes in the operating environment, traditional turbine controllers face many challenges in handling complex working conditions, especially when there are time lags and feedback delays.

[0003] In recent years, with the rapid development of information technology, control algorithms based on time series analysis have gradually become a research direction for the optimization of intelligent turbine controllers. Through time series analysis algorithms, the time lag characteristics, working characteristics and load changes of the system can be effectively captured, so as to make timely predictions on the response of the controller and optimize the control strategy; however, most of the existing technologies rely on a single feedback control strategy and lack effective compensation for time lag and feedback delay, resulting in control performance that is difficult to meet accuracy requirements under high dynamic response and complex working conditions.

[0004] In the existing intelligent turbine control technology, the system is usually adjusted based on feedback control, but its effect is limited when dealing with complex situations such as system time lag, load fluctuation and external interference; traditional time lag compensation methods are mostly based on static models and cannot adapt to the dynamic changes of the system in real time; in addition, the adjustment of the feedback control amount is only based on the difference between the real-time error and the set value, which cannot effectively capture the potential time lag characteristics in the system, resulting in the time lag effect cannot be effectively predicted and compensated, affecting the overall response speed and stability of the system.

[0005] Therefore, when faced with highly dynamic operating conditions, intelligent turbine controllers often have problems such as insufficient control accuracy and long response delays, which restricts their application in high-efficiency systems. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: based on the time series analysis algorithm, the working characteristics and delay characteristics of the intelligent turbine controller are analyzed to construct a time lag prediction model; The time-lag prediction model is used to calculate the feedforward compensation amount, and the feedback control amount is obtained based on the error adjustment between the target output and the actual output; Combining the feedback control amount and the feedforward compensation amount to obtain a time lag compensation amount; real-time monitoring of the operating data of the intelligent turbine controller, and dynamically adjusting the time delay compensation amount according to the control system state; The time delay differences of the multiple intelligent turbine controllers are individually compensated through a multi-machine time delay compensation mechanism, and the control signals of the intelligent turbine controllers are synchronously adjusted in combination with a global optimal time delay compensation strategy.

[0009] As a preferred solution of the collaborative control method for the intelligent turbine controller of the present invention, the control input signal, feedback signal and target output signal of the intelligent turbine controller are collected by the sensor module at a sampling period of once per second on the time axis; By cleaning, filling missing values, removing noise and normalizing the collected control input signal, feedback signal and target output signal, all data are converted to a unified range, dimension differences are eliminated, and the normalized data is formed into time series data through timestamp alignment.

[0010] As a preferred solution of the collaborative control method for an intelligent turbine controller described in the present invention, the time series data is expressed as a sequence of signal values ​​arranged in chronological order, each sampling point corresponds to a timestamp, and the corresponding signal value is collected at each time point.

[0011] As a preferred solution of the collaborative control method for the intelligent turbine controller described in the present invention, the control input signal time series is: ; The feedback signal time series is ; The target output signal time series is: ; Among them, n is the sequence number.

[0012] As a preferred embodiment of the collaborative control method for the intelligent turbine controller of the present invention, the preprocessed control input signal and feedback signal The composed multi-dimensional time series data is used as input, and the LSTM unit layer is selected to capture the timing characteristics in the historical control signal, predict the time lag between the future output and the control signal, and output the time lag prediction value; The time lag prediction value includes the time lag and time lag error.

[0013] As a preferred solution of the collaborative control method for the intelligent turbine controller of the present invention, obtaining the time delay compensation amount includes: Calculating a control error by comparing an actual output value of a feedback signal with a target output value of the target output signal; Calculating a feedforward compensation amount by combining the time lag prediction value and the control error; The feedback control amount is obtained based on the error adjustment between the target output and the actual output; The feedforward compensation amount and the feedback control amount are added to obtain the time lag compensation amount.

[0014] As a preferred solution of the collaborative control method for the intelligent turbine controller described in the present invention, the mathematical expression formula of the time delay compensation amount is: in, is the feedforward compensation, is the feedback control quantity.

[0015] As a preferred solution of the collaborative control method for the intelligent turbine controller described in the present invention, the control signals of each intelligent turbine controller are synchronously adjusted in combination with the global optimal time-delay compensation strategy, including: Each controller calculates the local compensation amount according to the current error and time lag, and then adjusts the compensation amount of each controller synchronously according to the global error and the gradient descent algorithm until the convergence condition is met: If the system error The change is less than the threshold in K consecutive iterations , the system is considered to have converged; After convergence, stop optimization adjustment and output the final control signal; That is, when the system reaches a convergent state, the final global compensation is output: Each controller will globally compensate As the adjustment quantity, it is fed back to the system to adjust the respective control signals to ensure that the system as a whole tends to the optimal state.

[0016] Beneficial effects of the present invention: 1. By using the time series analysis algorithm, the modeling of the working characteristics of the intelligent turbine controller and the accurate capture of the delay characteristics are realized. By collecting and analyzing the historical data of the controller, the dynamic response characteristics of the system are clarified and the possible time lag changes are predicted. By building a time lag prediction model, the time lag is predicted more accurately, avoiding control failure and system instability caused by hysteresis reaction, so that the control system can respond to dynamic conditions more flexibly and take compensation measures in advance, thereby improving the response speed and stability of the system; 2. The feedforward compensation adjusts the control input in advance based on the predicted time lag, making the system respond more timely and reducing the instability caused by the lag. The feedback control effectively compensates for the impact caused by the difference between the control signal and the actual output by calculating the error in real time and adjusting it, ensuring that the system output is always close to the expected value. The combination of feedforward and feedback not only improves the accuracy of the control system, but also enables the system to maintain stable operation when time lag occurs, and can more accurately respond to control signal lag and external disturbances, greatly improving the performance and efficiency of the intelligent turbine controller. 3. It can eliminate the impact caused by environmental fluctuations, ensure that the control system always maintains the best response characteristics, and avoid instability or performance degradation caused by delayed adjustment of the system; 4. Under the collaborative work of multiple machines, the system can eliminate the performance inconsistency caused by time lag differences between controllers through collaborative control. Through global optimization adjustment, each controller can work under the condition of minimum global error, thereby improving the overall coordination and stability of the system. Through global collaborative optimization, the advantages of each controller are fully utilized to avoid local optimal solutions in the system, thereby improving the overall control efficiency and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 It is a flow chart of the collaborative control method for the intelligent turbine controller shown in the present invention; Figure 2 It is a response time comparison schematic diagram shown in the present invention; Figure 3 It is a schematic diagram of system error comparison shown in the present invention; Figure 4 This is a schematic diagram for comparing the control accuracy shown in the present invention. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0019] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without making any creative work should fall within the scope of protection of the present invention.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] The existing intelligent turbine control technology has certain deficiencies in time-delay compensation and feedback delay, and cannot effectively solve the response delay problem caused by the system's time-delay characteristics, affecting the system's control accuracy and response speed. The present invention solves the problem of insufficient time-delay compensation in traditional control methods through time-delay prediction and feedforward feedback control strategies based on time series analysis algorithms, and optimizes the compensation amount of each controller through a multi-machine collaborative control strategy, thereby achieving the global optimal control effect of the system.

[0022] According to an embodiment of the present invention, Figure 1 The flowchart shown in the figure is a collaborative control method for an intelligent turbine controller, which specifically includes the following steps: S1. Analyze the working characteristics and delay characteristics of the intelligent turbine controller based on the time series analysis algorithm and build a time lag prediction model; S2, using the time-lag prediction model to calculate the feedforward compensation amount, and obtaining the feedback control amount based on the error adjustment between the target output and the actual output; S3, combining the feedback control amount and the feedforward compensation amount to obtain a time lag compensation amount; S4, real-time monitoring of the operation data of the intelligent turbine controller, and dynamically adjusting the time delay compensation amount according to the control system state; S5. The time delay differences of the multiple intelligent turbine controllers are individually compensated through a multi-machine time delay compensation mechanism, and the control signals of the intelligent turbine controllers are synchronously adjusted in combination with a global optimal time delay compensation strategy.

[0023] It should be noted that existing intelligent turbine controllers usually face the complexity of time lag, delay and multi-machine collaborative control. Traditional control systems have limited capabilities in dealing with time lag and multi-machine collaboration, and are prone to reduced control accuracy due to response delays, and it is difficult to ensure the coordination of multiple controllers. The present invention, by conducting an in-depth analysis of the working characteristics of the intelligent turbine controller based on a time series analysis algorithm, and combining time lag prediction models, a combination of feedforward and feedback control, real-time data monitoring and multi-machine collaborative optimization, can effectively overcome the time lag problem in the prior art and achieve accurate time lag compensation and multi-machine collaborative optimization.

[0024] The following describes in more detail the implementation process and / or effects of certain embodiments of the present invention in conjunction with some preferred or optional examples of the present invention.

[0025]

Building a time-lag prediction model

Get the time delay compensation amount

Multi-machine collaborative optimization control

[0026] It should also be noted that in modern wind power generation systems, intelligent turbine controllers are used to adjust the working state of turbine units to ensure the stability and efficiency of system operation. However, due to wind speed fluctuations, load changes, and response delays of the control system, traditional control methods are usually difficult to effectively compensate for time delays, resulting in slow system response and performance degradation. Therefore, this embodiment proposes a collaborative control method for an intelligent turbine controller, and conducts a comparative experiment with the traditional method to verify the obvious advantages of the present invention in time delay compensation and multi-machine collaborative control. The common control method of traditional turbine control system is that PID controller adjusts the control signal by adjusting the proportional, integral and differential parameters. However, since PID control cannot accurately predict and compensate for the time lag of the system, its performance often has large errors in complex dynamic environments. Specific defects include: the inability to dynamically adjust the control parameters to cope with the change of time lag, the inability to effectively compensate for the time lag difference when multiple machines work together, and the unstable control effect when the load fluctuates or the external interference is large, resulting in slow system response; The method of the present invention adopts a time-delay prediction model based on time series analysis, monitors the operation data of the controller in real time, and dynamically compensates for the time delay by combining feedforward compensation and feedback control. In addition, through the multi-machine time-delay compensation mechanism, the time-delay differences of multiple controllers can be effectively compensated separately, and the control signals of each controller can be adjusted synchronously to achieve global optimal control. (1) Experimental equipment Intelligent turbine controller: The intelligent controller of modern wind turbine generator sets has real-time data acquisition and processing capabilities, and can control input parameters such as wind speed and load; Data acquisition system: used to collect the input and output signals of the controller in real time, including wind speed, load, speed, and power output; Computer and analysis software: The computer runs MATLAB software, combines the time series analysis algorithm of the present invention to predict the time lag, adjust the feedforward compensation amount and the feedback control amount, and the computer connects the controller and the data acquisition system to transmit data in real time; Test equipment: wind speed simulator (simulates fluctuations of different wind speeds), load change simulator (simulates system load fluctuations); (2) Experimental environment Ambient temperature: 25°C, keep the laboratory constant temperature to avoid interference of external environmental changes on experimental data; Wind speed fluctuation range: 5-25 m / s (simulating different wind speed environments); System load change: 10%-80% (simulating load change conditions); Data sampling time: collect data every 5 seconds; Experimental period: 10 hours, analyzed through multiple data sampling; (3) Experimental process Before the experiment begins, collect the controller's input signals (such as wind speed, load changes, etc.) and output signals (such as power output, speed, etc.) data, and record the delay information at each time point; Using the traditional method (PID control method): Applying the PID controller in the wind turbine generator set, setting the proportional, integral, and differential parameters to control the wind speed and load changes. The PID controller does not consider the dynamic changes of time lag and adopts a static control strategy; The method of the present invention (time delay prediction and multi-machine collaborative compensation) is adopted: based on the time series analysis algorithm, a time delay prediction model is constructed to predict the delay of the control system, the feedforward compensation amount is calculated and combined with the feedback control amount to perform time delay compensation, the controller operation data is monitored in real time, and the time delay compensation amount is dynamically adjusted according to the system state. In the case of multi-machine collaborative control, the time delay difference of each controller is compensated, and the global control signal is adjusted synchronously; Experimental indicators: system response time, error value and system stability; The experiment evaluates the performance of the control method by recording the system's response time, error value, and control accuracy index. Figure 2 , Figure 3 and Figure 4 , showing the experimental results of the traditional PID control method and the method of the present invention, where: Reference Figure 2 The response time of the traditional PID control method is relatively long, about 10 seconds on average, while the method of the present invention significantly shortens the response time, maintaining it at about 5 seconds. That is, the combination of the lag compensation mechanism and the feedforward control significantly improves the response speed of the control system. Reference Figure 3 , the fluctuation range of the system error under the PID control method is relatively large, maintained at about 5%, while the error of the method of the present invention is always kept within 1% through time lag prediction and real-time adjustment, showing an obvious control accuracy advantage; Reference Figure 4 The control accuracy of the method of the present invention is significantly higher than that of the traditional PID method. The accuracy of the traditional PID control method is stable at about 80%, while the accuracy of the method of the present invention is always maintained at more than 98%, indicating that the present invention has stronger stability and accuracy in a changing environment; Preferably, compared with the traditional PID control method, the method of the present invention can provide significant advantages in time lag compensation and multi-machine collaborative control. Through dynamic adjustment of time lag prediction, feedforward and feedback control, the method of the present invention effectively improves control accuracy, reduces system error, and shortens response time. In addition, the application of the present invention in multi-machine collaborative control effectively compensates for the time lag differences between multiple controllers and improves the performance and stability of the overall control system.

[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A collaborative control method for an intelligent turbine controller, characterized in that: include: Analyze the working characteristics and delay characteristics of the intelligent turbine controller based on the time series analysis algorithm and build a time-delay prediction model; The time-lag prediction model is used to calculate the feedforward compensation amount, and the feedback control amount is obtained based on the error adjustment between the target output and the actual output; Combining the feedback control amount and the feedforward compensation amount to obtain a time lag compensation amount; real-time monitoring of the operating data of the intelligent turbine controller, and dynamically adjusting the time delay compensation amount according to the control system state; The time delay differences of the multiple intelligent turbine controllers are individually compensated through a multi-machine time delay compensation mechanism, and the control signals of the intelligent turbine controllers are synchronously adjusted in combination with a global optimal time delay compensation strategy.

2. The collaborative control method for an intelligent turbine controller according to claim 1, characterized in that: Collecting the control input signal, feedback signal and target output signal of the intelligent turbine controller through the sensor module at a sampling period of once per second on the time axis; By cleaning, filling missing values, removing noise and normalizing the collected control input signal, feedback signal and target output signal, all data are converted to a unified range, dimension differences are eliminated, and the normalized data is formed into time series data through timestamp alignment.

3. The collaborative control method for an intelligent turbine controller according to claim 2, characterized in that: The time series data is expressed as a sequence of signal values ​​arranged in chronological order, each sampling point corresponds to a timestamp, and a corresponding signal value is collected at each time point.

4. The collaborative control method for an intelligent turbine controller according to claim 3, characterized in that: The control input signal time series is ; The feedback signal time series is ; The target output signal time series is: ; Among them, n is the sequence number.

5. The collaborative control method for an intelligent turbine controller according to claim 4, characterized in that: The control input signal after preprocessing and feedback signal The composed multi-dimensional time series data is used as input, and the LSTM unit layer is selected to capture the timing characteristics in the historical control signal, predict the time lag between the future output and the control signal, and output the time lag prediction value; The time lag prediction value includes the time lag and time lag error.

6. The collaborative control method for an intelligent turbine controller according to claim 1, characterized in that: Obtaining the time delay compensation amount includes: Calculating a control error by comparing an actual output value of a feedback signal with a target output value of the target output signal; Calculating a feedforward compensation amount by combining the time lag prediction value and the control error; The feedback control amount is obtained based on the error adjustment between the target output and the actual output; The feedforward compensation amount and the feedback control amount are added to obtain the time lag compensation amount.

7. The collaborative control method for an intelligent turbine controller according to claim 1, characterized in that: The mathematical expression formula of the time delay compensation amount is: in, is the feedforward compensation, is the feedback control quantity.

8. The collaborative control method for an intelligent turbine controller according to claim 1, characterized in that: The control signals of the intelligent turbine controllers are synchronously adjusted in combination with the global optimal time-delay compensation strategy, including: Each controller calculates the local compensation amount according to the current error and time lag, and then adjusts the compensation amount of each controller synchronously according to the global error and the gradient descent algorithm until the convergence condition is met: If the system error The change is less than the threshold in K consecutive iterations , the system is considered to have converged; After convergence, stop optimization adjustment and output the final control signal; That is, when the system reaches a convergent state, the final global compensation is output: Each controller will globally compensate As the adjustment quantity, it is fed back to the system to adjust the respective control signals to ensure that the system as a whole tends to the optimal state.

Images