Fan control optimization method and system based on Beidou receiver

Abstract

The invention relates to a fan control optimization method and system based on a Beidou receiver. The method is suitable for the technical field of fan control optimization. According to the technical scheme, the method comprises the steps that first operation wind energy information, first simulation wind energy information and a first simulation fan control strategy are obtained, and the first simulation wind energy information is obtained through prediction of a wind energy information prediction large model; comparing the first operation wind energy information with first simulation wind energy information corresponding to the first operation wind energy information in space-time, and generating first deviation information; deriving power output deviation caused by each wind energy deviation characteristic value in the first deviation information in the first simulation fan control strategy, and adding the wind energy deviation characteristic values with the power output deviation greater than a preset value into a characteristic set with correlation degree; updating the first simulation wind energy information based on the first operation wind energy information and the feature set with the correlation degree, and generating second simulation wind energy information; and inputting the second simulated wind energy information and the first simulated fan control strategy into the large power optimization model, and outputting a fan optimization control strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine control optimization, and in particular to a wind turbine control optimization method and system based on a Beidou receiver. Background Art

[0002] At present, wind power forecasting and scheduling technology has made certain progress and applications, but there are still some problems and challenges. Due to the randomness, uncertainty and nonlinearity of wind power changes, it is difficult for existing wind power forecasting models to accurately capture the laws and characteristics of wind power changes, thus affecting the accuracy and reliability of the forecast results. Summary of the invention

[0003] The technical problem to be solved by the present invention is: in response to the above-mentioned problems, a wind turbine control optimization method based on Beidou receiver is provided to solve the technical problem that the information coupling in the wind turbine control power optimization is complex and it is difficult to output accurate optimization results.

[0004] The technical solution adopted by the present invention is: a wind turbine control optimization method based on Beidou receiver, characterized in that it includes:

[0005] Acquire first operating wind energy information, first simulated wind energy information corresponding to the first operating wind energy information in time and space, and a first simulated wind turbine control strategy corresponding to the first simulated wind energy information, wherein the first simulated wind energy information is predicted by a large wind energy information prediction model;

[0006] Comparing the first operating wind energy information with the first simulated wind energy information corresponding to the first operating wind energy information in time and space to generate first deviation information;

[0007] In the first simulated wind turbine control strategy, the power output deviation caused by each wind energy deviation characteristic value in the first deviation information is derived, and the wind energy deviation characteristic value whose power output deviation is greater than a preset value is added to the characteristic set with a correlation degree;

[0008] Based on the first operating wind energy information and the feature set with a correlation, the first simulated wind energy information is updated to generate second simulated wind energy information;

[0009] The second simulated wind energy information and the first simulated wind turbine control strategy are input into the power optimization model, and the wind turbine optimization control strategy is output.

[0010] The obtaining of the first operating wind energy information includes:

[0011] The operation data and multi-dimensional monitoring data of the wind turbine generator set are acquired, and the operation data and the multi-dimensional monitoring data are processed to determine the first operation wind energy information.

[0012] The obtaining of the first simulated wind energy information and the first simulated wind turbine control strategy corresponding to the first simulated wind energy information includes:

[0013] Based on the environmental parameters at time t, a deep neural network model is used to perform wind energy information prediction, generate the first simulated wind energy information at time t+1 with time and space as dimensions, and generate the first simulated wind turbine control strategy corresponding to the first simulated wind energy information.

[0014] Before the second simulated wind energy information is generated, the method further includes:

[0015] Acquire second deviation information, compare the first operating fan control strategy with the first simulated fan control strategy, and generate second deviation information;

[0016] The data set of the wind turbine control strategy corresponding to the second deviation information is used to perform data filling on the first simulated wind energy information according to a feature set with a correlation degree, and then the second simulated wind energy information is generated.

[0017] The wind turbine control strategy includes: pitch angle adjustment, yaw angle adjustment and / or brake control.

[0018] A wind turbine control optimization system based on Beidou receiver, characterized by comprising:

[0019] An information acquisition module, used to acquire first operating wind energy information, first simulated wind energy information corresponding to the first operating wind energy information in time and space, and a first simulated wind turbine control strategy corresponding to the first simulated wind energy information, wherein the first simulated wind energy information is predicted by a large wind energy information prediction model;

[0020] An information comparison module, used for comparing the first operating wind energy information with the first simulated wind energy information corresponding to the first operating wind energy information in time and space, and generating first deviation information;

[0021] A correlation feature extraction module, used to derive, in the first simulated wind turbine control strategy, a power output deviation caused by each wind energy deviation feature value in the first deviation information, and add a wind energy deviation feature value whose power output deviation is greater than a preset value to a feature set with a correlation degree;

[0022] An information updating module, used to update the first simulated wind energy information based on the first operating wind energy information and the feature set with correlation, and generate the second simulated wind energy information;

[0023] The strategy optimization module is used to input the second simulated wind energy information and the first simulated wind turbine control strategy into the power optimization model and output the wind turbine optimization control strategy.

[0024] A device, characterized in that

[0025] include:

[0026] Memory for storing computer programs;

[0027] A processor is used to implement the steps of the wind turbine control optimization method based on Beidou receiver according to any one of claims 1 to 5 when executing the computer program.

[0028] A computer-readable storage medium storing a computer program, characterized in that:

[0029] When the computer program is executed by a processor, it is used to implement the steps of the wind turbine control optimization method based on a Beidou receiver according to any one of claims 1 to 5.

[0030] The beneficial effects of the present invention are as follows: the present invention simulates wind energy information to obtain wind energy modeling information, and on the basis of the wind energy modeling information, compares the actual wind energy operation conditions, analyzes the deviation information between the wind energy and the modeling information, and further processes the modeling based on the deviation information, so as to obtain efficient wind energy control decision information optimization.

[0031] The present invention can use deviation information to mark characteristic value information, so as to mark the characteristic information in the deviation information that has the greatest impact on the power output result, and use the above-mentioned first marking information to perform second marking information on the empirical results that have been run in the standard library and the power output model, so that the above-mentioned information can be associated, so that the characteristic value with the highest correlation with the power optimization output is marked from the characteristic value information, and the influence of the above-mentioned characteristic value on the adjustment of the wind turbine control strategy is analyzed, and the result of efficient output power optimization output is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flowchart of an embodiment. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] Embodiment 1: This embodiment discloses one implementation of a wind turbine control optimization method based on a Beidou receiver, and the optimization method comprises the following steps:

[0035] S1. Obtaining operation data and multi-dimensional monitoring data of the wind turbine generator set, processing the operation data and the multi-dimensional monitoring data, obtaining current actual first operation wind energy information, and a first operation wind turbine control strategy and corresponding power output corresponding to the first operation wind energy information.

[0036] Using a deep neural network model, wind energy information prediction modeling is performed to generate first simulated wind energy information with time and space as dimensions and corresponding to the time and space of the first operating wind energy information, and generate a first simulated wind turbine control strategy corresponding to the first simulated wind energy information.

[0037] In this embodiment, the wind turbine control strategy includes pitch angle adjustment, yaw angle adjustment and / or stall control to maximize the output power of the wind turbine.

[0038] S2. Compare the first operating wind energy information and the first simulated wind energy information, and generate first deviation information according to the comparison result of the two.

[0039] S3. In the first simulated wind turbine control strategy, the power output deviation caused by each wind energy deviation characteristic value in the first deviation information is derived, and the wind energy deviation characteristic value whose power output deviation is greater than the preset value is added to the feature set with correlation.

[0040] S4. Based on the first operating wind energy information and the feature set with a correlation, update the first simulated wind energy information to generate second simulated wind energy information.

[0041] Obtain deviation information between the first simulated wind energy information and the current operating wind energy information, perform matching retrieval in the first operating wind energy information, obtain a feature set with a correlation corresponding to the deviation information, adjust and update the feature set with a correlation in the first operating wind turbine control strategy accordingly, and generate second simulated wind energy information.

[0042] S5. Input the second simulated wind energy information and the first simulated wind turbine control strategy into the power optimization model, and output the wind turbine optimization control strategy.

[0043] In this embodiment, the deviation information between the first simulated wind energy information and the current operating wind energy information is obtained by: using an error evaluation algorithm, calculating the deviation between the simulated data and the actual data, and classifying the deviation results, so as to calculate the influence of the error information on the execution result;

[0044] A first mark is performed on the deviation information in the first simulated wind energy information, a second mark is performed on the associated information corresponding to the first mark in the first operating wind energy information, and a correlation feature set analysis is performed on the first marked information to obtain a current feature set.

[0045] The first mark is: inserting deviation data into the deviation information in the first simulated wind energy information to obtain the training result of the first operating wind energy information;

[0046] The second mark is: the training result of the first operation wind energy information is analyzed, and the second mark is executed with the wind turbine control strategy adjustment item that meets the normal operation conditions in the first operation wind energy information;

[0047] Perform feature set analysis based on the feature set generated by the second tag to analyze the correlation conclusion, perform adjustment strategy analysis based on the feature set with correlation, and obtain the optimal result output.

[0048] The feature set operation with correlation in the first running wind turbine control strategy is correspondingly adjusted and updated to obtain the current wind turbine control strategy.

[0049] The analysis of executing adjustment strategy of feature set with correlation is as follows: executing training with feature set with correlation as adjustment condition, and taking the first simulated fan control strategy as input of power optimization model, obtaining the output result of the model, thereby adjusting the first simulated fan control strategy to obtain the current fan control strategy.

[0050] The specific steps of generating the feature set with correlation are: deriving the power output information caused by the deviation information in the first simulated wind turbine control strategy, and selecting the feature value whose power output result is greater than a certain threshold as the feature set with correlation.

[0051] Embodiment 2: This embodiment discloses a wind turbine control optimization system based on Beidou receiver, including:

[0052] An information acquisition module, used to acquire first operating wind energy information, first simulated wind energy information corresponding to the first operating wind energy information in time and space, and a first simulated wind turbine control strategy corresponding to the first simulated wind energy information, wherein the first simulated wind energy information is predicted by a large wind energy information prediction model;

[0053] An information comparison module, used for comparing the first operating wind energy information with the first simulated wind energy information corresponding to the first operating wind energy information in time and space, and generating first deviation information;

[0054] A correlation feature extraction module, used to derive, in the first simulated wind turbine control strategy, a power output deviation caused by each wind energy deviation feature value in the first deviation information, and add a wind energy deviation feature value whose power output deviation is greater than a preset value to a feature set with a correlation degree;

[0055] An information updating module, used to update the first simulated wind energy information based on the first operating wind energy information and the feature set with correlation, and generate the second simulated wind energy information;

[0056] The strategy optimization module is used to input the second simulated wind energy information and the first simulated wind turbine control strategy into the power optimization model and output the wind turbine optimization control strategy.

[0057] Embodiment 3: This embodiment is a specific implementation of Embodiment 1. In the above implementation, a wind turbine control optimization method based on a Beidou receiver is disclosed.

[0058] The above method mainly includes the following steps:

[0059] The current wind speed measurement information obtained is input into a wind energy simulation prediction model to obtain first simulated wind energy information. The simulated wind energy information mainly obtains predicted wind energy information within a certain time segment. The wind energy prediction model is such as ultra-short-term, short-term or medium- and long-term predictions known to those skilled in the art, usually combined with numerical weather forecasting (NWP), wind power conversion model, statistical learning method, machine learning technology, etc. The wind energy information includes but is not limited to: wind speed, wind direction, wind speed distribution, wind power density, wind and other analysis of wind energy information itself;

[0060] The current wind speed measurement information is subjected to a data conversion process to obtain first operating wind energy information in the same format as the wind energy simulation prediction large model data output;

[0061] Acquire first deviation information, compare the first operating wind energy information with the first simulated wind energy information, and thus acquire prediction deviation information;

[0062] Obtaining second deviation information, inputting the first operating wind energy information and the first simulated wind energy information into the power optimization model at the same time, and obtaining deviation information on the power optimization control strategy; the above deviation information comes from the first operating wind turbine control strategy and the first simulated wind turbine control strategy;

[0063] Analyze and obtain the characteristic value information of wind energy from the first deviation information, that is, what kind of changes have occurred in the actual value of the wind energy information compared with the predicted information, and first determine the degree of influence on the power output when these actual values ​​change, that is: obtain the characteristic association information in the wind turbine control strategy associated with the wind turbine operating parameters in the changed wind energy characteristic information, and perform a first mark on the above wind energy characteristic value information in the first simulated wind energy information, the first mark includes information on the degree of influence, and the characteristic association information in the wind turbine control strategy associated with it;

[0064] According to the first tag information and the first deviation information, a search is performed in the first simulated wind turbine control strategy, and the relationship data between the wind energy information and the power output corresponding to the first deviation information is collected within the deviation range. For example, the wind speed corresponding to a certain position point in the first deviation information will have a certain deviation. What is the power data outputted by the power optimization large model under the wind speed range information, and the control strategy eigenvalue matrix set under the corresponding power optimization output condition;

[0065] A corresponding second mark is made for the above-mentioned control strategy eigenvalue matrix set, and the control strategy change caused by the change of wind energy information in the control strategy eigenvalue matrix set is obtained from the second deviation information. The second mark is supplemented and corrected, and the wind energy information data corresponding to the corresponding supplemented and corrected information is collected, and the correlation information in the first simulated wind energy information is filled with data to form the second simulated wind energy information. In this way, the second simulated wind energy information after the data is filled is input into the power optimization model to output the current wind turbine control strategy.

[0066] The wind turbine control strategy according to the present invention includes but is not limited to: Maximum power point tracking (MPPT) strategy: According to real-time information such as wind speed and wind direction, the speed and blade angle of the wind turbine are dynamically adjusted to keep it in the maximum power output state. Yaw control: Using advanced wind direction prediction algorithms, the yaw angle of the wind turbine is adjusted in advance so that the wind wheel is always facing the incoming wind direction, thereby improving the efficiency of wind energy capture. Turbulence control: In areas where wind speed and wind direction vary greatly, the operating parameters of the wind turbine are adjusted to reduce the impact of turbulence on the wind turbine and ensure the stable operation of the wind turbine. High-precision sensors: High-precision sensors are used to monitor the operating status and environmental parameters of the wind turbine in real time to provide accurate data support for the control system. Fault diagnosis and prediction: Using fault diagnosis and prediction technology, potential faults of the wind turbine are discovered in advance, and corresponding maintenance measures are taken to reduce fault downtime and improve the operating efficiency of the wind turbine. Optimization scheduling strategy: According to the overall operation of the wind farm, a reasonable wind turbine scheduling strategy is formulated to ensure that the wind turbine operates in the best state. Load balancing: By balancing the load of wind turbines, the load pressure of a single unit is reduced, and the operating efficiency of the entire wind farm is improved. Multi-unit coordinated control: Realize the coordinated control of multiple wind turbines to improve the wind energy capture capability and operating efficiency of the entire wind farm.

[0067] In one specific implementation, the adjustment strategy is to optimize the power of the wind turbine generator set by adjusting and controlling the wind turbine's pitch control, yaw control, generator parameter control, etc. This method can improve the wind energy conversion efficiency of the wind turbine generator set, achieve maximum wind energy capture, and thus optimize the power control of the wind turbine generator set.

[0068] In one specific implementation, by optimizing the power curve of the wind turbine generator set, the torque and power of the wind turbine generator set change relatively slowly within the operating wind speed range, reducing the flicker size of the wind turbine generator and achieving a better control effect, thereby improving the power control accuracy and stability of the wind turbine generator.

[0069] According to one of the control strategy steps of the power output model of the present invention, the specific implementation method is as follows:

[0070] By establishing an accurate wind speed prediction model, the future wind speed change trend can be predicted, so as to adjust the operating parameters of the wind turbine in advance, optimize the power output, and dynamically adjust the power-wind speed (Pv) curve of the wind turbine according to the wind speed changes to adapt to different wind speed conditions. For example, under low wind speed conditions, the normal operation of the wind turbine can be maintained by changing the additional power provided by the motor excitation source. The pitch angle of the blades can be adjusted using variable pitch technology to change the wind energy utilization coefficient and achieve maximum power capture at different wind speeds. Below the rated wind speed, the generator torque and impeller speed can be controlled to maintain the optimal tip speed ratio and improve energy conversion efficiency. According to the distribution characteristics of wind direction and wind speed, the spatial layout of wind turbines in the wind farm can be optimized to reduce the wake effect and improve the power output efficiency of the entire wind farm. Advanced control strategies such as model predictive control can be used to dynamically adjust the operating status of each wind turbine in the wind farm according to real-time data of wind speed and wind direction to achieve optimal distribution of overall power. Improved genetic algorithms can be used to optimize the dispatching strategy of the wind farm, taking into account factors such as wind speed changes, wind turbine characteristic curves and grid demand to maximize the overall output power of the wind farm. The power output adjustment methods include: controlling the active power output of the wind farm to adapt to the needs and stability requirements of the grid and reduce the impact of wind power grid connection on the stability and security of the power system.

[0071] The error evaluation method according to the present invention includes but is not limited to: root mean square error (RMSE), mean absolute error (MAE), and deviation model.

[0072] According to the system and method of the present invention, the wind farm can respond to the changes of wind speed and wind direction more flexibly and efficiently, and optimize the power output. By obtaining the change of the wind rotor acceleration and correspondingly controlling the wind rotor blade pitch angle to optimize the adjustment, a high utilization rate of wind energy can be achieved during the wind speed change process. This method can improve the power stability and wind energy utilization efficiency of the wind turbine.

[0073] As a further illustration of the present invention, the control of a wind turbine is a complex process involving multiple aspects and multiple physical field coupling technologies. The technical solution of the present invention is based on the simulation and calculation between the feature sets associated with the model and may not include all situations involving control strategy adjustments.

[0074] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A wind turbine control optimization method based on Beidou receiver, characterized in that: include: Acquire first operating wind energy information, first simulated wind energy information corresponding to the first operating wind energy information in time and space, and a first simulated wind turbine control strategy corresponding to the first simulated wind energy information, wherein the first simulated wind energy information is predicted by a large wind energy information prediction model; Comparing the first operating wind energy information with the first simulated wind energy information corresponding to the first operating wind energy information in time and space to generate first deviation information; In the first simulated wind turbine control strategy, the power output deviation caused by each wind energy deviation characteristic value in the first deviation information is derived, and the wind energy deviation characteristic value whose power output deviation is greater than a preset value is added to the characteristic set with a correlation degree; Based on the first operating wind energy information and the feature set with a correlation, the first simulated wind energy information is updated to generate second simulated wind energy information; The second simulated wind energy information and the first simulated wind turbine control strategy are input into the power optimization model, and the wind turbine optimization control strategy is output.

2. The wind turbine control optimization method based on Beidou receiver according to claim 1 is characterized in that: The obtaining of the first operating wind energy information includes: The operation data and multi-dimensional monitoring data of the wind turbine generator set are acquired, and the operation data and the multi-dimensional monitoring data are processed to determine the first operation wind energy information.

3. The wind turbine control optimization method based on Beidou receiver according to claim 1 is characterized in that: The obtaining of the first simulated wind energy information and the first simulated wind turbine control strategy corresponding to the first simulated wind energy information includes: Based on the environmental parameters at time t, a deep neural network model is used to perform wind energy information prediction, generate the first simulated wind energy information at time t+1 with time and space as dimensions, and generate the first simulated wind turbine control strategy corresponding to the first simulated wind energy information.

4. The wind turbine control optimization method based on Beidou receiver according to claim 1 is characterized in that: Before the second simulated wind energy information is generated, the method further includes: Acquire second deviation information, compare the first operating fan control strategy with the first simulated fan control strategy, and generate second deviation information; The data set of the wind turbine control strategy corresponding to the second deviation information is used to perform data filling on the first simulated wind energy information according to a feature set with a correlation degree, and then the second simulated wind energy information is generated.

5. The wind turbine control optimization method based on Beidou receiver according to claim 1 is characterized in that: The wind turbine control strategy includes: pitch angle adjustment, yaw angle adjustment and / or brake control.

6. A wind turbine control optimization system based on Beidou receiver, characterized in that: include: An information acquisition module, used to acquire first operating wind energy information, first simulated wind energy information corresponding to the first operating wind energy information in time and space, and a first simulated wind turbine control strategy corresponding to the first simulated wind energy information, wherein the first simulated wind energy information is predicted by a large wind energy information prediction model; An information comparison module, used for comparing the first operating wind energy information with the first simulated wind energy information corresponding to the first operating wind energy information in time and space, and generating first deviation information; A correlation feature extraction module, used to derive, in the first simulated wind turbine control strategy, a power output deviation caused by each wind energy deviation feature value in the first deviation information, and add a wind energy deviation feature value whose power output deviation is greater than a preset value to a feature set with a correlation degree; An information updating module, used to update the first simulated wind energy information based on the first operating wind energy information and the feature set with correlation, and generate the second simulated wind energy information; The strategy optimization module is used to input the second simulated wind energy information and the first simulated wind turbine control strategy into the power optimization model and output the wind turbine optimization control strategy.

7. A device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the wind turbine control optimization method based on Beidou receiver according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the wind turbine control optimization method based on a Beidou receiver according to any one of claims 1 to 5.

Images