Wind power cluster multi-station reactive voltage cooperative control device and method

Abstract

The invention relates to the technical field of voltage control, in particular to a wind power cluster multi-station reactive voltage cooperative control device and method, and the device comprises an AVC main station which is responsible for the overall voltage optimization control of a system; the AVC coordinated control substation is located between the AVC master station and the AVC substations, receives the voltage control instruction sent by the AVC master station, converts the voltage control instruction into a reactive instruction and issues the reactive instruction to each AVC substation for execution; the AVC substations are used for receiving reactive power instructions issued by the AVC coordinated control substations, controlling reactive power sources in respective areas and carrying out closed-loop control on voltage target values; according to the parameter control system, through two sets of control parameters, the control requirements of the same application scene are met, and switching of the control parameters is carried out. According to the invention, the problems of voltage oscillation and large control difficulty caused by large response time discreteness in the prior art are effectively solved, the response speed of the system and the stability of voltage control are improved, and different wind power plant layouts and control requirements are met.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage control, and in particular to a device and method for coordinated reactive voltage control of multiple stations in a wind power cluster. Background Art

[0002] Wind power generation, as a renewable energy source, has been widely used around the world. Wind farms convert wind energy into electrical energy through wind turbines and transmit it to the power grid. However, due to the instability and unpredictability of wind energy, the access of wind farms to the power grid will bring challenges to the voltage stability and reactive power balance of the power grid. In order to cope with these problems, existing technologies usually adopt automatic voltage control (AVC) systems.

[0003] At present, the voltage control of wind farm groups mainly adopts the AVC master station-AVC substation two-layer control architecture. In this architecture, the AVC master station is responsible for coordinating the voltage control of the entire wind farm group and sending voltage control instructions to each AVC substation. Each substation controls the reactive power source in its own responsible area according to these instructions to achieve the target voltage. However, in the application scenario of wind farm groups with short electrical distances, the response time of different substation systems is highly discrete, which can easily cause severe voltage fluctuations and there is a risk of non-convergence.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a wind power cluster multi-station reactive power voltage coordinated control device and system, thereby effectively solving the problems pointed out in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: A wind power cluster multi-station reactive power voltage coordinated control device, comprising: AVC master station, responsible for the overall voltage optimization control of the system; The AVC co-control substation is located between the AVC master station and the AVC substation, receives the voltage control instruction sent by the AVC master station, converts the voltage control instruction into a reactive power instruction, and sends it to each AVC substation for execution; The AVC substation receives the reactive power instruction issued by the AVC co-control substation, controls the reactive power sources in the respective areas, and performs closed-loop control of the voltage target value; The parameter control system is connected with the AVC main station, the AVC co-control substation and the AVC substation through data communication. Through two sets of control parameters, it is respectively applicable to the control requirements of different application scenarios, and the control parameters are switched to affect the voltage optimization control strategy of the AVC main station, adjust the way in which the AVC co-control substation generates reactive power instructions, and change the control strategy of the AVC substation for reactive power sources.

[0007] Furthermore, it also includes: an adaptive learning system, which optimizes control parameters and adjustment strategies by analyzing historical data. The adaptive learning system is connected to the AVC master station, AVC co-control substation, AVC substation and parameter control system, and adaptively adjusts control parameters based on historical operation data collected from each of the AVC substations. The optimized control instructions are issued to the AVC master station through the parameter control system, and the generation method of reactive power instructions is adjusted through the AVC co-control substation.

[0008] Furthermore, by analyzing historical data, control parameters and adjustment strategies are optimized, including: Obtain voltage, reactive power and related control parameters during system operation and establish a historical database; Performing system stability influence identification and pattern recognition on the historical data in the historical database to obtain historical data analysis results; Obtaining a prediction model based on the historical data analysis results and a machine learning algorithm; According to the prediction model, the voltage fluctuation and reactive power demand in the future operating environment are predicted, and the control strategy is adjusted in advance.

[0009] Furthermore, the machine learning algorithm is a gradient boosting decision tree.

[0010] Furthermore, the AVC co-control substation includes: A parsing module, used for receiving a control target calculated by the AVC master station and parsing the control target into a voltage target value; A signal module outputs a voltage increase blocking signal, a voltage reduction blocking signal or a bidirectional voltage blocking signal according to the current voltage real-time value and the voltage upper and lower limits; A collection module, collecting the real-time voltage of the controlled bus and the reactive power; A mode selection module enables a corresponding control mode according to the control word setting; A fixed value algorithm module is used to set control parameters and calculate reactive power control target values ​​according to the input voltage target value and actual voltage conditions; The instruction distribution module decomposes reactive power instructions according to the calculated reactive power control target value, and sends the instructions to each AVC substation for execution.

[0011] Furthermore, the fixed value algorithm module includes: The fixed value unit includes the voltage regulation step, cycle, dead zone and reactive power regulation step, cycle, dead zone of the wind farm grid connection point; The algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the parameters of the setting unit.

[0012] Furthermore, the AVC assisting substation also includes: a redundant control module, which automatically takes over all functions of the main control module when the main control module fails. The main control module is composed of an analysis module, a signal module, an acquisition module, a mode selection module, a constant value algorithm module and an instruction allocation module.

[0013] Furthermore, the redundant control module includes: A redundant parsing unit receives and parses the control target from the AVC master station. When the parsing module fails, the redundant parsing unit can automatically take over the parsing task and correctly convert the control target; The redundant signal processing unit is responsible for processing the current real-time voltage value and the voltage upper and lower limit information. When the signal module fails, it generates a voltage increase lock signal, a voltage reduction lock signal or a bidirectional voltage lock signal; The redundant mode selection unit is used to enable the corresponding control mode according to the control word setting, and can be automatically activated to switch and execute the control mode when the mode selection module fails; The redundant fixed value algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the control parameters preset in the fixed value algorithm module when the fixed value algorithm module fails, and sends the calculation result to the instruction distribution module; The redundant instruction distribution unit, when the instruction distribution module fails, decomposes the reactive instruction according to the reactive target value output by the redundant algorithm unit, and sends the instruction to each AVC substation for execution.

[0014] A method for coordinated reactive voltage control of multiple stations in a wind power cluster, using the above-mentioned device for coordinated reactive voltage control of multiple stations in a wind power cluster, comprising: The AVC master station calculates the overall optimization target and sends voltage control instructions; The auxiliary control substation receives the voltage control instruction and parses the voltage control instruction into a specific reactive power control instruction; Collect bus voltage and reactive power data in real time, and calculate reactive power target value based on voltage target and real-time data; Each of the AVC substations receives and executes a reactive power target value, and performs closed-loop control of a voltage target value.

[0015] Furthermore, the method also includes: optimizing control parameters and adjustment strategies by analyzing historical data.

[0016] The technical solution of the present invention can achieve the following technical effects: The invention solves the problem of voltage oscillation and control difficulty caused by large discreteness of response time in the prior art, improves the response speed of the system and the stability of voltage control, and adapts to different wind farm layouts and control requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a framework diagram of a wind power cluster multi-station reactive power and voltage coordinated control device; Figure 2 Schematic diagram of the process for optimizing control parameters and adjustment strategies; Figure 3 It is a flow chart of a method for coordinated reactive power and voltage control of multiple stations in a wind power cluster; DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0021] Embodiment 1 like Figure 1 As shown, the present invention provides a wind power cluster multi-station reactive power voltage coordinated control device, the device comprising: AVC master station, responsible for the overall voltage optimization control of the system; The AVC auxiliary control substation is located between the AVC master station and the AVC substation. It receives the voltage control instructions sent by the AVC master station, converts the voltage control instructions into reactive power instructions, and sends them to each AVC substation for execution; The AVC substation receives reactive power instructions issued by the AVC cooperative control substation, controls reactive power sources in their respective areas, and performs closed-loop control of voltage target values; Parameter control system, parameter control system, is connected with AVC main station, AVC co-control substation and AVC substation through data communication. Through two sets of control parameters, it is suitable for the control requirements of different application scenarios, and switches the control parameters to affect the voltage optimization control strategy of AVC main station, adjust the way AVC co-control substation generates reactive power instructions, and change the control strategy of AVC substation for reactive power source.

[0022] Specifically, the AVC master station calculates the voltage demand of the power grid, generates voltage control instructions, and sends these instructions to the AVC co-control substation to coordinate the reactive voltage regulation of the wind farm group. The AVC master station uses optimization algorithms to analyze the real-time status of the power grid, determine the voltage target value of each wind farm, and ensure the stable operation of the entire power grid. Based on the optimization calculation results, the AVC master station passes the voltage control instructions to the AVC co-control substation, providing a basis for subsequent reactive power control.

[0023] The AVC cooperative control substation is located between the AVC master station and the AVC substation. It receives the voltage control instructions sent by the AVC master station and parses these instructions into specific reactive power control instructions. The cooperative control substation selects appropriate control parameters and algorithms according to the grid operation mode (such as two-layer control mode or three-layer control mode). The parsed reactive power control instructions will be sent to each AVC substation to ensure that each wind farm can coordinate and coordinate reactive power voltage regulation.

[0024] The AVC substation receives reactive power control instructions from the AVC co-control substation, and controls the reactive sources (such as capacitor banks, sensors, etc.) in their respective areas according to these instructions to achieve closed-loop control of the voltage target value. In the process of executing reactive power control, the AVC substation monitors and feeds back the voltage and reactive power data in the area in real time to ensure that the control effect meets the expectations.

[0025] The parameter control system sets two sets of control parameters, which are suitable for different application scenarios, such as two-layer control mode and three-layer control mode. Under different operating modes, the parameter control system can dynamically switch control parameters to adapt to different power grid operation requirements. By optimizing and adjusting the control parameters, the parameter control system can improve the response speed and stability of the system and ensure the accuracy and reliability of reactive voltage regulation.

[0026] The present invention effectively solves the problems of voltage oscillation and control difficulty caused by large discreteness of response time in the prior art, improves the response speed of the system and the stability of voltage control, and adapts to different wind farm layouts and control requirements.

[0027] As a preferred embodiment of the above embodiment, it also includes: an adaptive learning system, which optimizes control parameters and adjustment strategies by analyzing historical data, and the adaptive learning system is connected to the AVC master station, AVC co-control substation, AVC substation and parameter control system, and adaptively adjusts control parameters based on historical operation data collected from each of the AVC substations, and issues optimized control instructions to the AVC master station through the parameter control system, and adjusts the generation method of reactive power instructions through the AVC co-control substation.

[0028] Specifically, the adaptive learning system uses machine learning algorithms to continuously adjust and optimize control parameters based on historical operating data and real-time feedback data to ensure that the system can operate in the best state under different operating conditions. By training and analyzing data under different working conditions, the adaptive learning system can predict future operating trends and adjust control strategies in advance, thereby improving the response speed and stability of the system and further reducing voltage fluctuations and reactive power losses.

[0029] As a preferred embodiment of the above, Figure 2 As shown in the figure, by analyzing historical data, the control parameters and adjustment strategies are optimized, including: A10: Obtain the voltage, reactive power and related control parameters during system operation and establish a historical database; A20: Perform system stability influence identification and pattern recognition on historical data in the historical database to obtain historical data analysis results; A30: Obtain a prediction model based on historical data analysis results and machine learning algorithms; A40: Based on the prediction model, predict the voltage fluctuation and reactive power demand in the future operating environment, and adjust the control strategy in advance.

[0030] Specifically, first, during the operation of the system, data on voltage, reactive power and related control parameters are collected in real time, and these data are continuously stored in the historical database to form a comprehensive historical operation data record. These data provide detailed information about the system under different operating conditions.

[0031] Next, the data stored in the historical database is analyzed, including system stability influence identification and pattern recognition. By analyzing the impact of different factors on system stability in historical data, we can identify which parameter changes have a significant impact on voltage stability and reactive power control. At the same time, we use pattern recognition technology to identify the operating modes and features in historical data and find out the typical behavior of the system under different operating conditions.

[0032] Based on the above analysis results, a prediction model is established using a machine learning algorithm. First, the historical data is cleaned and normalized to remove noise data and standardize the data format; then the preprocessed data is used to train the machine learning model so that it can identify the impact of different parameter changes on system operation. During the training process, the accuracy of the model is verified by using some historical data, and the model parameters are adjusted to improve the prediction accuracy.

[0033] Using the trained prediction model, the currently collected voltage, reactive power and related control parameters are input in real time to predict the voltage fluctuation and reactive power demand in the future operating environment. The model calculates the future voltage change trend and reactive power demand based on the input data, and adjusts the control strategy in advance according to the prediction results. For example, if it is predicted that the voltage will fluctuate in the future, the system can adjust the reactive power control parameters in advance to stabilize the voltage change.

[0034] As a preferred embodiment of the above embodiment, the machine learning algorithm is a gradient boosting decision tree.

[0035] The specific reasons why this solution uses the gradient boosted decision tree (GBDT) as the machine learning algorithm are as follows: Adapting to complex voltage control requirements: In the coordinated reactive voltage control of multiple stations in a wind power cluster, the changes in voltage and reactive power are affected by multiple factors, including the layout of the wind farm, real-time weather conditions, load fluctuations, etc. There are complex nonlinear relationships between these factors. GBDT can effectively capture these complex nonlinear relationships and provide accurate prediction models to ensure that the system can identify voltage fluctuations in advance and make adjustments.

[0036] High-dimensional data processing capability: In this solution, a large amount of real-time monitoring data and historical operation data need to be processed. These data include multiple features such as voltage, reactive power, and control parameters. GBDT performs well in processing high-dimensional data and can automatically select and combine the most important features to improve the prediction accuracy and efficiency of the model.

[0037] Robustness and stability: The operating environment of wind farms is highly uncertain, and there may be noise and missing values ​​in the data. GBDT is highly robust to noise and missing values, and can provide reliable prediction results under imperfect data conditions. By continuously optimizing control parameters and strategies, GBDT can improve system stability and reduce voltage fluctuations and reactive power losses.

[0038] Model interpretability: In this solution, the AVC master station and the AVC cooperative control substation need to adjust the control strategy according to the prediction results. The results of GBDT have good interpretability, which can help engineers understand the impact of various features on voltage control and optimize the control strategy. For example, by analyzing the importance of each feature in the model, it is possible to identify which factors have the greatest impact on voltage fluctuations, so as to make targeted adjustments.

[0039] As a preferred embodiment of the above, the AVC co-control substation includes: A parsing module, used for receiving the control target calculated by the AVC master station and parsing the control target into a voltage target value; A signal module outputs a voltage increase blocking signal, a voltage decrease blocking signal or a bidirectional voltage blocking signal according to the current voltage real-time value and the voltage upper and lower limits; The acquisition module collects the real-time voltage and reactive power of the controlled bus; A mode selection module enables a corresponding control mode according to the control word setting; The fixed value algorithm module is used to set the control parameters and calculate the reactive power control target value according to the input voltage target value and the actual voltage situation; The instruction distribution module decomposes the reactive power instructions according to the calculated reactive power control target value and sends the instructions to each AVC substation for execution.

[0040] Specifically, the analysis module receives voltage control instructions from the AVC master station, and converts the control target into specific voltage target values ​​through a preset analysis algorithm. These voltage target values ​​are used to guide subsequent reactive power control operations; the signal module monitors the current voltage value in real time, and when the voltage exceeds the set upper and lower limits, it outputs a corresponding blocking signal to protect the safe operation of the system. For example, when the voltage is too high, it outputs a voltage increase blocking signal, and when the voltage is too low, it outputs a voltage reduction blocking signal. In some cases, it outputs a bidirectional voltage blocking signal; the acquisition module uses sensors and monitoring equipment to collect voltage and reactive power data of the controlled bus in real time, and feeds this data back to the analysis module and the constant value algorithm module for further calculation and control; the mode selection module selects the voltage and reactive power data of the controlled bus in real time according to the operation requirements of the power grid. and control word settings, select to enable double-layer control mode or three-layer control mode. The double-layer control mode is suitable for simpler grid structures and has a faster response speed. The three-layer control mode is suitable for complex grid structures and has a higher control accuracy. The constant value algorithm module sets the corresponding control parameters, and calculates the reactive power control target value using a preset algorithm (such as PID control algorithm) based on the input voltage target value and real-time collected voltage data. These target values ​​will be used to guide the reactive power control operation of the AVC substation. The instruction distribution module decomposes the reactive power control target values ​​calculated by the constant value algorithm module into specific reactive power control instructions, and distributes them to each AVC substation for execution, ensuring that each wind farm can coordinate and coordinate reactive voltage regulation to achieve the overall voltage control target of the system.

[0041] As a preferred embodiment of the above, the fixed value algorithm module includes: The constant value unit is used to set and store a series of basic control parameters, including voltage regulation step, cycle, dead zone, and reactive power regulation step, cycle, dead zone; The algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the parameters of the setting unit.

[0042] Specifically, the fixed value algorithm module includes a fixed value unit and an algorithm unit. The fixed value unit is used to set and store a series of basic control parameters, including voltage regulation step, cycle, dead zone, and reactive regulation step, cycle, and dead zone. These parameters are set by the system during initialization and stored in the fixed value unit for use in the control process. The algorithm unit is responsible for calculating the reactive target value based on the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the parameters of the fixed value unit. The algorithm unit combines the voltage target value with the actual collected voltage and reactive power data, and uses a preset control algorithm, such as a PID control algorithm, to dynamically calculate the reactive control target value, taking into account parameters such as the regulation step, cycle, and dead zone, to ensure that the voltage operates stably within the target range. Through the collaborative work of the fixed value unit and the algorithm unit, the fixed value algorithm module can achieve accurate parameter setting and reactive target value calculation, ensuring the stability and response speed of the system under different operating conditions.

[0043] As a preferred embodiment of the above embodiment, the AVC assisting substation also includes: a redundant control module, which automatically takes over all functions of the main control module when the main control module fails. The main control module includes an analysis module, a signal module, an acquisition module, a mode selection module, a constant value algorithm module and an instruction allocation module.

[0044] As a preferred embodiment of the above, the redundant control module includes: The redundant parsing unit receives and parses the control targets from the AVC master station. When the parsing module fails, the redundant parsing unit can automatically take over the parsing task and correctly convert the control targets. The redundant signal processing unit is responsible for processing the current real-time voltage value and the voltage upper and lower limit information. When the signal module fails, it generates a voltage increase lock signal, a voltage reduction lock signal or a bidirectional voltage lock signal; The redundant mode selection unit is used to enable the corresponding control mode according to the control word setting, and can be automatically activated to switch and execute the control mode when the mode selection module fails; The redundant fixed value algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the control parameters preset in the fixed value algorithm module when the fixed value algorithm module fails, and sends the calculation result to the instruction distribution module; The redundant instruction distribution unit, when the instruction distribution module fails, decomposes the reactive instruction according to the reactive target value output by the redundant algorithm unit, and sends the instruction to each AVC substation for execution.

[0045] Specifically, the redundant analysis unit is responsible for receiving and analyzing the control targets from the AVC master station. When the analysis module fails, the redundant analysis unit can automatically take over the analysis task, correctly convert the control targets, and ensure that the generation of voltage target values ​​is not interrupted; the redundant signal processing unit is responsible for processing the current real-time voltage value and the upper and lower voltage limit information. When the signal module fails, the redundant signal processing unit can generate a voltage increase lock signal, a voltage reduction lock signal or a bidirectional voltage lock signal to ensure the safe operation of the system under abnormal voltage conditions; the redundant mode selection unit enables the corresponding control mode according to the control word setting. When the mode selection module fails, the redundant mode selection unit can automatically activate, switch and execute the control mode, and ensure the correct selection of the system operation mode; when the fixed value algorithm module fails, the redundant fixed value algorithm unit can calculate the reactive target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the control parameters preset in the fixed value algorithm module, and send the calculation result to the instruction distribution module to ensure the accurate calculation of the reactive control target value; when the instruction distribution module fails, the redundant instruction distribution unit can decompose the reactive instruction according to the reactive target value output by the redundant fixed value algorithm unit, and send the instruction to each AVC substation for execution, to ensure that each wind farm coordinates and coordinates the reactive voltage regulation. Through the configuration of the redundant control module, even if the main control module fails, the AVC cooperative control substation can still maintain normal operation to ensure the stability and reliability of the system.

[0046] Embodiment 2 like Figure 3 As shown, the present application provides a method for coordinated reactive voltage control of multiple stations in a wind power cluster, which adopts the above-mentioned device for coordinated reactive voltage control of multiple stations in a wind power cluster, and includes: S1: The AVC master station calculates the overall optimization target and sends voltage control instructions; S2: The auxiliary control substation receives the voltage control instruction and parses the voltage control instruction into a specific reactive power control instruction; S3: collect bus voltage and reactive power data in real time, and calculate reactive power target value based on voltage target and real-time data; S4: Each AVC substation receives and executes the reactive power target value, and performs closed-loop control of the voltage target value. As a preferred embodiment of the above embodiment, the method further includes: optimizing control parameters and adjustment strategies by analyzing historical data.

[0047] The technical effects achieved by this embodiment are the same as those of the first embodiment and will not be described again here.

[0048] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined herein and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A wind power cluster multi-station reactive power voltage coordinated control device, characterized in that: include: AVC master station, responsible for the overall voltage optimization control of the system; The AVC co-control substation is located between the AVC master station and the AVC substation, receives the voltage control instruction sent by the AVC master station, converts the voltage control instruction into a reactive power instruction, and sends it to each AVC substation for execution; The AVC substation receives the reactive power instruction issued by the AVC co-control substation, controls the reactive power sources in the respective areas, and performs closed-loop control of the voltage target value; The parameter control system is connected with the AVC main station, the AVC co-control substation and the AVC substation through data communication. Through two sets of control parameters, it is respectively applicable to the control requirements of different application scenarios, and the control parameters are switched to affect the voltage optimization control strategy of the AVC main station, adjust the way in which the AVC co-control substation generates reactive power instructions, and change the control strategy of the AVC substation for reactive power sources.

2. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 1 is characterized in that: Also includes: An adaptive learning system, wherein the adaptive learning system optimizes control parameters and adjustment strategies by analyzing historical data. The adaptive learning system is connected to the AVC master station, the AVC co-control substation, the AVC substation and the parameter control system. Based on the historical operation data collected from each of the AVC substations, the control parameters are adaptively adjusted, and the optimized control instructions are issued to the AVC master station through the parameter control system. The generation method of reactive power instructions is adjusted through the AVC co-control substation.

3. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 2 is characterized in that: By analyzing historical data, we can optimize control parameters and adjustment strategies, including: Obtain voltage, reactive power and related control parameters during system operation and establish a historical database; Performing system stability influence identification and pattern recognition on the historical data in the historical database to obtain historical data analysis results; Obtaining a prediction model based on the historical data analysis results and a machine learning algorithm; According to the prediction model, the voltage fluctuation and reactive power demand in the future operating environment are predicted, and the control strategy is adjusted in advance.

4. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 3 is characterized in that: The machine learning algorithm is a gradient boosted decision tree.

5. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 1, characterized in that: The AVC cooperative control substation includes: A parsing module, used for receiving a control target calculated by the AVC master station and parsing the control target into a voltage target value; A signal module outputs a voltage increase blocking signal, a voltage reduction blocking signal or a bidirectional voltage blocking signal according to the current voltage real-time value and the voltage upper and lower limits; A collection module, collecting the real-time voltage of the controlled bus and the reactive power; A mode selection module, which enables a corresponding control mode according to the control word setting, wherein the control mode affects the calculation method of the reactive power control target value and the instruction allocation strategy; A fixed value algorithm module is used to set control parameters and calculate reactive power control target value according to the input voltage target value and actual voltage conditions, wherein the actual voltage conditions include the blocking signal output by the signal module and the real-time voltage and reactive power collected by the collection module; The instruction distribution module decomposes reactive power instructions according to the calculated reactive power control target value, and sends the instructions to each AVC substation for execution.

6. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 5, characterized in that: The fixed value algorithm module includes: The constant value unit is used to set and store a series of basic control parameters, including voltage regulation step, cycle, dead zone, and reactive power regulation step, cycle, dead zone; The algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the parameters of the setting unit.

7. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 1, characterized in that: The AVC assisting substation also includes: a redundant control module, which automatically takes over all functions of the main control module when a main control module fails. The main control module includes a parsing module, a signal module, an acquisition module, a mode selection module, a constant value algorithm module and an instruction distribution module.

8. The wind power cluster multi-station reactive power and voltage coordinated control device according to claim 7, characterized in that: The redundant control module comprises: A redundant parsing unit receives and parses the control target from the AVC master station. When the parsing module fails, the redundant parsing unit can automatically take over the parsing task and correctly convert the control target; The redundant signal processing unit is responsible for processing the current real-time voltage value and the voltage upper and lower limit information. When the signal module fails, it generates a voltage increase lock signal, a voltage reduction lock signal or a bidirectional voltage lock signal; The redundant mode selection unit is used to enable the corresponding control mode according to the control word setting, and can be automatically activated to switch and execute the control mode when the mode selection module fails; The redundant fixed value algorithm unit calculates the reactive power target value according to the voltage target value output by the analysis module, the bus voltage value and reactive power output by the acquisition module, and the control parameters preset in the fixed value algorithm module when the fixed value algorithm module fails, and sends the calculation result to the instruction distribution module; The redundant instruction distribution unit, when the instruction distribution module fails, decomposes the reactive instruction according to the reactive target value output by the redundant algorithm unit, and sends the instruction to each AVC substation for execution.

9. A method for coordinated reactive voltage control of multiple stations in a wind power cluster, which is adopted in the device for coordinated reactive voltage control of multiple stations in a wind power cluster as claimed in claims 1-8, characterized in that: include: The AVC master station calculates the overall optimization target and sends voltage control instructions; The auxiliary control substation receives the voltage control instruction and parses the voltage control instruction into a specific reactive power control instruction; Collect bus voltage and reactive power data in real time, and calculate reactive power target value based on voltage target and real-time data; Each of the AVC substations receives and executes a reactive power target value, and performs closed-loop control of a voltage target value.

10. The method for coordinated reactive power and voltage control of multiple wind power stations in a wind power cluster according to claim 9, characterized in that: The method also includes: optimizing control parameters and adjustment strategies by analyzing historical data.

Images