CN115456046A

CN115456046A - Wind driven generator speed and pitch changing control system and method based on active disturbance rejection control

Info

Publication number: CN115456046A
Application number: CN202211014641.3A
Authority: CN
Inventors: 叶林; 王力军; 陈思; 梁哲铭; 史祥; 刘美岑; 李春廷; 杨小龙; 刘铭
Original assignee: Huaneng Renewables Corp Ltd; Beijing Huaneng Xinrui Control Technology Co Ltd
Current assignee: Huaneng Renewables Corp Ltd; Beijing Huaneng Xinrui Control Technology Co Ltd
Priority date: 2022-08-23
Filing date: 2022-08-23
Publication date: 2022-12-09

Abstract

The application relates to the field of intelligent control of wind power generators, and particularly discloses a variable speed and variable pitch control system and method of a wind power generator based on active disturbance rejection control. Therefore, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, and the wind power conversion efficiency is controlled to protect the fan from being damaged.

Description

Wind driven generator speed and pitch changing control system and method based on active disturbance rejection control

Technical Field

The invention relates to the field of intelligent control of wind driven generators, in particular to a variable speed and variable pitch control system and a method thereof of a wind driven generator based on active disturbance rejection control.

Background

With the increasing permeability of wind energy to the power grid, the influence of the wind power generation system on the frequency and voltage stability of the power grid is more and more obvious. The pitch system is a complex nonlinear system disturbed by various uncertainties and is a key component of the wind power generation system.

When the wind speed exceeds the rated cut-in wind speed, the pitch angle is changed to control the wind power conversion efficiency, thereby capturing rated power from the wind and protecting the wind turbine from damage, i.e., when the wind speed is higher than the rated speed of the variable speed variable pitch wind turbine, the pitch angle is changed to maintain the output power and the rotor speed at their rated values. From the point of view of grid integration, the power control technology of wind turbine generator sets is also becoming more and more important.

However, in the control process of the existing pitch control system, although fluctuation of output power can be reduced, the fluctuation amplitude can not meet application requirements under complex working conditions, and grid fluctuation during grid connection is caused. The reason for this is that existing pitch systems adjust the pitch angle based on the wind conditions at the current point in time, and the moment of adjustment is already the next moment, and the wind conditions at the next moment have changed. That is, conventional pitch systems have a weak control hysteresis effect, causing the waveforms they produce to exacerbate grid fluctuations.

Therefore, a more optimized variable speed and variable pitch control scheme of the wind driven generator is expected.

Disclosure of Invention

The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a variable speed and variable pitch control system and a method thereof for a wind driven generator based on active disturbance rejection control, which adopt an artificial intelligence control technology, extract multi-scale neighborhood correlation of wind speed values and pitch angles of a plurality of preset time points including the current time point and dynamic change characteristics of output power of the wind driven generator through a multi-scale neighborhood characteristic extraction module respectively, dynamically predict data so as to complete variable speed and variable pitch control of the wind driven generator, and introduce a Bayesian model to perform predictive adjustment in the process so as to enable the predicted result to be more accurate. Therefore, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, and the wind power conversion efficiency is controlled to protect the fan from being damaged.

According to an aspect of the application, a variable speed and variable pitch control system of a wind driven generator based on active disturbance rejection control is provided, and comprises:

the data acquisition module is used for acquiring wind speed values of a plurality of preset time points including the current time point, pitch angles of the plurality of preset time points including the current time point and output power of the wind driven generator of the plurality of preset time points including the current time point, wherein the wind speed values are acquired by a wind speed sensor;

the wind speed data coding module is used for arranging the wind speed values of the plurality of preset time points including the current time point into a wind speed input vector and then obtaining a multi-scale wind speed feature vector through the first multi-scale neighborhood feature extraction module;

the pitch angle data encoding module is used for arranging the pitch angles of the plurality of preset time points including the current time point into a pitch angle input vector and then obtaining a multi-scale pitch angle feature vector through the second multi-scale neighborhood feature extraction module;

the power data coding module is used for arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into a power input vector and then obtaining a multi-scale power feature vector through the third multi-scale neighborhood feature extraction module;

a Bayesian fusion module for fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain a posterior probability vector;

the posterior information correction module is used for correcting the characteristic values of all the positions in the posterior probability vector based on the mean value and the variance of the characteristic value sets of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and

and the control result generation module is used for enabling the corrected posterior probability vector to pass through a classifier to obtain a classification result, and the classification result is used for indicating that the pitch angle of the current time point should be increased or decreased.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the wind speed data encoding module includes: the wind speed vectorization unit is used for arranging the wind speed values of a plurality of preset time points including the current time point into the wind speed input vector; a wind speed first scale convolution coding unit, configured to input the wind speed input vector into a first convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale wind speed associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a wind speed second scale convolution coding unit, configured to input the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale wind speed associated feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the wind speed multi-scale characteristic cascading unit is used for cascading the first neighborhood scale wind speed associated characteristic vector and the second neighborhood scale wind speed associated characteristic vector to obtain the multi-scale wind speed characteristic vector.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the pitch angle data encoding module includes: a pitch angle vectorization unit, configured to arrange the pitch angles of the multiple predetermined time points including the current time point as the pitch angle input vector; a pitch angle first scale convolution coding unit, configured to input the pitch angle input vector into a first convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale pitch angle associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length; a pitch angle second scale convolution encoding unit, configured to input the pitch angle input vector into a second convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale pitch angle associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and the pitch angle multi-scale feature cascading unit is used for cascading the first neighborhood scale pitch angle association feature vector and the second neighborhood scale pitch angle association feature vector to obtain the multi-scale pitch angle feature vector.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the power data encoding module includes: the power vectorization unit is used for arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into the power input vector; a power first scale convolution coding unit, configured to input the power input vector into a first convolution layer of the third multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale power correlation feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a power second scale convolution coding unit, configured to input the power input vector into a second convolution layer of the third multi-scale neighborhood feature extraction module to obtain a second neighborhood scale power correlation feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the power multi-scale feature cascading unit is used for cascading the first neighborhood scale power correlation feature vector and the second neighborhood scale power correlation feature vector to obtain the multi-scale power feature vector.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the bayesian fusion module is further configured to: fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain the posterior probability vector;

wherein the formula is:

qi＝pi*ai/bi

where pi is the eigenvalue of each position in the multi-scale pitch angle eigenvector, ai and bi are the eigenvalues of each position in the multi-scale power eigenvector and the multi-scale wind speed eigenvector, respectively, and qi is the eigenvalue of each position in the a posteriori probability vector.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the posterior information correction module is further configured to: based on the mean and variance of the feature value sets of all positions in the posterior probability vector, correcting the feature values of all positions in the posterior probability vector by the following formula to obtain the corrected posterior probability vector;

wherein the formula is:

wherein v is _i The characteristic value of each position in the posterior probability vector is represented, the characteristic value of each position in the corrected posterior probability vector is represented, mu and sigma are characteristic set v _i E.v, V represents the posterior probability vector, exp (-) represents the exponential operation of the vector, the exponential operation raised to the power of the vector represents the natural exponential function value raised to the power of the value at each position of the vector.

In the above wind turbine variable speed and variable pitch control system based on active disturbance rejection control, the control result generation module is further configured to: processing the corrected a posteriori probability vector using the classifier to obtain the classification result with the formula: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) I X }, wherein W ₁ To W _n As a weight matrix, B ₁ To B _n Is a bias vector and X is the corrected posterior probability vector.

According to another aspect of the application, a wind turbine variable speed and variable pitch control method based on active disturbance rejection control comprises the following steps:

acquiring wind speed values of a plurality of preset time points including a current time point, pitch angles of the plurality of preset time points including the current time point and output power of the wind driven generator of the plurality of preset time points including the current time point, wherein the wind speed values are acquired by a wind speed sensor;

arranging the wind speed values of the plurality of preset time points including the current time point as a wind speed input vector, and then obtaining a multi-scale wind speed feature vector through a first multi-scale neighborhood feature extraction module;

after the pitch angles of the multiple preset time points including the current time point are arranged as pitch angle input vectors, a multi-scale pitch angle feature vector is obtained through a second multi-scale neighborhood feature extraction module;

arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into a power input vector, and then obtaining a multi-scale power feature vector through a third multi-scale neighborhood feature extraction module;

fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain a posterior probability vector;

correcting the characteristic values of all positions in the posterior probability vector based on the mean value and the variance of the characteristic value sets of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and

and passing the corrected posterior probability vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the pitch angle of the current time point should be increased or decreased.

In the above method for controlling variable speed and variable pitch of a wind turbine based on active disturbance rejection control, after arranging the wind speed values at a plurality of predetermined time points including the current time point as a wind speed input vector, obtaining a multi-scale wind speed feature vector through a first multi-scale neighborhood feature extraction module, the method includes: arranging the wind speed values of a plurality of preset time points including the current time point into the wind speed input vector; inputting the wind speed input vector into a first convolution layer of the first multi-scale neighborhood characteristic extraction module to obtain a first neighborhood scale wind speed associated characteristic vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; inputting the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale wind speed associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and cascading the first neighborhood scale wind speed associated feature vector and the second neighborhood scale wind speed associated feature vector to obtain the multi-scale wind speed feature vector.

In the above method for controlling speed and pitch variation of a wind turbine based on active disturbance rejection control, after arranging the pitch angles of the plurality of predetermined time points including the current time point as a pitch angle input vector, obtaining a multi-scale pitch angle feature vector by a second multi-scale neighborhood feature extraction module, the method includes: arranging the pitch angles of the plurality of preset time points including the current time point as the pitch angle input vector; inputting the pitch angle input vector into a first convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a first neighborhood region scale pitch angle correlation feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; inputting the pitch angle input vector into a second convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale pitch angle associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; and cascading the first neighborhood scale pitch angle associated feature vector and the second neighborhood scale pitch angle associated feature vector to obtain the multi-scale pitch angle feature vector.

In the above wind turbine variable speed and variable pitch control method based on active disturbance rejection control, after arranging the output power of the wind turbine at a plurality of predetermined time points including the current time point as a power input vector, obtaining a multi-scale power feature vector through a third multi-scale neighborhood feature extraction module, the method includes: arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into the power input vector; inputting the power input vector into a first convolution layer of the third multi-scale neighborhood feature extraction module to obtain a first neighborhood scale power correlation feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; inputting the power input vector into a second convolution layer of the third multi-scale neighborhood region feature extraction module to obtain a second neighborhood region scale power correlation feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and cascading the first neighborhood scale power association feature vector and the second neighborhood scale power association feature vector to obtain the multi-scale power feature vector.

In the above method for controlling variable speed and variable pitch of a wind turbine based on active disturbance rejection control, fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a bayesian probability model to obtain an a posteriori probability vector, including: fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain the posterior probability vector;

wherein the formula is:

qi＝pi*ai/bi

In the above wind turbine variable speed and variable pitch control method based on active disturbance rejection control, based on the mean and variance of the eigenvalue sets of all positions in the posterior probability vector, the method corrects the eigenvalue of each position in the posterior probability vector to obtain a corrected posterior probability vector, and includes: based on the mean and variance of the feature value sets of all positions in the posterior probability vector, correcting the feature values of all positions in the posterior probability vector by the following formula to obtain the corrected posterior probability vector;

wherein the formula is:

In the above wind turbine variable speed and variable pitch control method based on active disturbance rejection control, passing the corrected posterior probability vector through a classifier to obtain a classification result, where the classification result is used to indicate that the pitch angle at the current time point should be increased or decreased, and includes: processing the corrected a posteriori probability vector using the classifier to obtain the classification result with the formula: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) I X }, wherein W ₁ To W _n As a weight matrix, B ₁ To B _n Is a bias vector and X is the corrected posterior probability vector.

Compared with the prior art, the active disturbance rejection control-based variable speed and variable pitch control system and the method thereof adopt an artificial intelligence control technology, extract the multi-scale neighborhood correlation of the wind speed values and the pitch angles of a plurality of preset time points including the current time point and the dynamic change characteristics of the output power of the wind driven generator through a multi-scale neighborhood characteristic extraction module respectively, dynamically predict data, and further complete variable speed and variable pitch control of the wind driven generator, and introduce a Bayesian model to perform predictive adjustment in the process, so that the predicted result is more accurate. Therefore, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, the wind power conversion efficiency is controlled, and the fan is protected from being damaged.

Drawings

The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.

Fig. 1 is an application scenario diagram of a variable speed and variable pitch control system of a wind turbine based on active disturbance rejection control according to an embodiment of the application.

FIG. 2 is a block diagram of a wind turbine variable speed and pitch control system based on active disturbance rejection control according to an embodiment of the application.

FIG. 3 is a block diagram of a wind turbine speed and pitch control system based on active disturbance rejection control, wherein the wind turbine speed and pitch control system is provided with a speed data encoding module.

FIG. 4 is a flowchart of a wind turbine variable speed and variable pitch control method based on active disturbance rejection control according to an embodiment of the application.

Fig. 5 is a schematic architecture diagram of a wind turbine variable speed and pitch control method based on active disturbance rejection control according to an embodiment of the application.

Detailed Description

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only a few embodiments of the present application, and not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.

Overview of scenes

As previously mentioned, as wind energy becomes more permeable to the grid, the impact of wind power generation systems on grid frequency and voltage stability becomes more and more significant. The pitch system is a complex nonlinear system disturbed by various uncertainties and is a key component of the wind power generation system.

When the wind speed exceeds the rated cut-in wind speed, the pitch angle is changed to control the wind power conversion efficiency, thereby capturing rated power from the wind and protecting the wind turbine from damage, i.e., when the wind speed is higher than the rated speed of the variable speed pitch wind turbine, the pitch angle is changed to maintain the output power and the rotor speed at their rated values. From the point of view of grid integration, the power control technology of wind turbine generator sets is also becoming more and more important.

However, in the control process of the existing pitch control system, although fluctuation of output power can be reduced, the fluctuation amplitude can not meet application requirements under complex working conditions, and grid fluctuation during grid connection is caused. The reason for this is that existing pitch systems adjust the pitch angle based on the wind conditions at the current point in time, and the moment of adjustment is already the next moment, and the wind conditions at the next moment have changed. That is, conventional pitch systems have a weak control hysteresis effect, causing the waveforms they produce to exacerbate grid fluctuations. Therefore, a more optimized variable speed and variable pitch control scheme of the wind driven generator is expected.

Accordingly, the present inventors considered that when a pitch system of a wind turbine is controlled to control wind power conversion efficiency to protect a wind turbine from damage, an existing pitch system is adjusted based on a wind condition at a current time point, so that an adjusted pitch angle cannot keep output power and rotor speed at rated values, and thus grid fluctuation is aggravated. Therefore, in the technical scheme of the application, data prediction is expected to be dynamically performed based on the wind speed values and the pitch angles of a plurality of preset time points including the current time point and the dynamic change characteristic information of the output power of the wind driven generator, so that the speed and pitch control of the wind driven generator is completed, and a Bayesian model is introduced in the process for predictive adjustment, so that the predicted result is more accurate, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, the wind power conversion efficiency is controlled, and the wind driven generator is protected from being damaged.

Specifically, in the technical solution of the present application, first, wind speed values at a plurality of predetermined time points including a current time point are collected by a wind speed sensor, pitch angles at the plurality of predetermined time points including the current time point are collected by an angle sensor, and output power of the wind power generator at the plurality of predetermined time points including the current time point is collected by a power detector.

Then, the convolutional neural network with excellent performance in local implicit association feature extraction is used for implicit feature extraction of data feature association. It should be understood that convolutional neural networks were originally models applied in the image domain, but the idea of local feature extraction can be applied to time series data analysis as well. For example, a time-series convolution structure with a convolution kernel size of 3, the convolution kernel is moved along the time dimension in the form of a sliding window for time-series data input, and outputs a weighted sum of the data within each time-series segment. Each convolution unit stacks a plurality of convolution kernels to output a multi-dimensional feature. The large convolution kernel can extract features from a large-scale time sequence neighborhood, wherein the influence of each item of numerical value in the neighborhood is smaller, so that the fluctuation of input data is weakened, and the influence of noise points on output features is reduced. However, the difference of numerical value changes is weakened by the large-scale convolution kernel, and the problem of excessive smoothness is easily caused, so that the output characteristics lose the discrimination capability. In contrast, small scale convolution kernels are better able to retain information in the input data, but are also more susceptible to interference from noise therein. Therefore, the convolution units with different sizes are combined to extract the features of different time sequence scales in consideration of the characteristics of convolution with different scales. And then completing feature fusion by adopting a feature splicing mode, thereby obtaining the multi-scale neighborhood features.

That is, specifically, in the technical solution of the present application, first, the wind speed values at the plurality of predetermined time points including the current time point are arranged as a wind speed input vector to facilitate subsequent feature extraction. Further, convolution layers with one-dimensional convolution kernels of different scales of the multi-scale neighborhood feature extraction module are used for respectively carrying out one-dimensional convolution coding on the wind speed input vector, and then the obtained feature vectors corresponding to the two one-dimensional convolution kernels of different scales are cascaded to obtain the multi-scale wind speed feature vector. Particularly, through the method, the multi-scale neighborhood correlation of the dynamic change characteristics of the wind speed values in the time sequence dimension is extracted, so that the output characteristics not only comprise the smoothed characteristics, but also keep the original input characteristics, the information loss is avoided, and the accuracy of subsequent classification is improved. In other examples of the present application, the multi-scale neighborhood feature extraction module may further include a greater number of one-dimensional convolution layers, which perform neighborhood-related feature extraction of different scales using one-dimensional convolution kernels of different lengths, which is not limited in this application.

Similarly, considering that the pitch angle at the plurality of predetermined time points including the current time point and the output power of the wind driven generator at the plurality of predetermined time points including the current time point also have dynamic implicit change characteristics in the time dimension, in order to extract the multi-scale neighborhood correlation of the dynamic change characteristics, the pitch angle at the plurality of predetermined time points including the current time point and the output power of the wind driven generator at the plurality of predetermined time points including the current time point are further respectively input into the multi-scale neighborhood characteristic extraction module to be processed so as to obtain a multi-scale pitch angle characteristic vector and a multi-scale power characteristic vector. Therefore, the output correlation characteristics in neighborhoods with different scales of the dynamic change characteristics of the pitch angle and the output power of the wind driven generator not only contain the characteristics after smoothing, but also store the original input characteristic information, so as to avoid the loss of the information and further improve the accuracy of subsequent classification.

It should be appreciated that in view of using the multi-scale pitch angle feature vector as a prior probability, the technical solution in the present application aims at updating the prior probability to get a posterior probability upon new evidence, i.e. upon a new change of the wind speed value. Then, according to the bayesian formula, the posterior probability is the prior probability multiplied by the event probability divided by the evidence probability, and therefore, in the technical solution of the present application, a bayesian probability model is used to fuse the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector to obtain the posterior probability vector, where the multi-scale pitch angle feature vector is used as the prior, the multi-scale wind speed feature vector is used as the event, and the multi-scale power feature vector is used as the evidence. In this way, the a posteriori probability vectors can be passed through the classifier to obtain a classification result indicating that the pitch angle at the current point in time should be increased or decreased.

However, since the bayesian probability model performs point-by-point calculation on the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector, the obtained posterior probability vector may have a problem of poor expression consistency among feature values, thereby affecting the classification effect.

Based on this, in the technical solution of the present application, the posterior probability vector is further subjected to recursive squeeze-excitation optimization in length dimension, which is expressed as:

wherein v is _i The characteristic value of each position in the posterior probability vector is represented, the characteristic value of each position in the corrected posterior probability vector is represented, mu and sigma are characteristic set v _i E.g., the mean and variance of V, V representing the a posteriori probability vector, exp (-) representing the exponential operation of the vector, the exponential operation raised to the vector representing the natural exponential function value raised to the power of the value at each position of the vector.

Here, the length-dimension recursive squeeze-excitation optimization activates the dimension recursion along the length direction of the feature distribution based on the statistical characteristics of the feature set, so as to infer the distribution of each sampling point of the feature in the length dimension thereof, and a predetermined-dimension squeeze-excitation mechanism composed of a ReLU-Sigmoid function is adopted to obtain a predetermined-dimension attention-enhanced confidence value, so as to improve the expression consistency of the high-dimension feature flow pattern of the posterior probability vector in the length dimension, and further improve the classification accuracy.

Based on this, this application has proposed a aerogenerator variable speed pitch control system based on active disturbance rejection control, it includes: the data acquisition module is used for acquiring wind speed values of a plurality of preset time points including the current time point, pitch angles of the plurality of preset time points including the current time point and output power of the wind driven generator of the plurality of preset time points including the current time point, wherein the wind speed values are acquired by a wind speed sensor; the wind speed data coding module is used for arranging the wind speed values of the plurality of preset time points including the current time point into a wind speed input vector and then obtaining a multi-scale wind speed feature vector through the first multi-scale neighborhood feature extraction module; the pitch angle data encoding module is used for arranging the pitch angles of the plurality of preset time points including the current time point into a pitch angle input vector and then obtaining a multi-scale pitch angle feature vector through the second multi-scale neighborhood feature extraction module; the power data coding module is used for arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into a power input vector and then obtaining a multi-scale power feature vector through the third multi-scale neighborhood feature extraction module; a Bayesian fusion module for fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain a posterior probability vector; the posterior information correction module is used for correcting the characteristic values of all the positions in the posterior probability vector based on the mean value and the variance of the characteristic value sets of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and the control result generation module is used for enabling the corrected posterior probability vector to pass through a classifier to obtain a classification result, and the classification result is used for indicating that the pitch angle of the current time point should be increased or decreased.

Fig. 1 illustrates an application scenario of a variable speed and variable pitch control system of a wind turbine based on active disturbance rejection control according to an embodiment of the application. As shown in fig. 1, in this application scenario, first, wind speed values at a plurality of predetermined time points including a current time point are collected by a wind speed sensor (e.g., W as illustrated in fig. 1), pitch angles at the plurality of predetermined time points including the current time point are collected by an angle sensor (e.g., a as illustrated in fig. 1) disposed in a pitch control system (e.g., P as illustrated in fig. 1), and output power of a wind power generator (e.g., G as illustrated in fig. 1) at the plurality of predetermined time points including the current time point is collected by a power detector (e.g., P as illustrated in fig. 1). Then, the obtained wind speed values, the pitch angles and the output power of the wind turbine at the plurality of predetermined time points including the current time point are input into a server (for example, a cloud server S as illustrated in fig. 1) deployed with an active disturbance rejection control-based wind turbine variable speed and pitch control algorithm, wherein the server can process the wind speed values, the pitch angles and the output power of the wind turbine at the plurality of predetermined time points including the current time point by the active disturbance rejection control-based wind turbine variable speed and pitch control algorithm to generate a classification result indicating that the pitch angle at the current time point should be increased or decreased.

In this application scenario, the declaration registration information includes, but is not limited to, "operating status", "wireless IP address", "cluster priority", and "cluster controller", etc.

Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.

Exemplary System

FIG. 2 illustrates a block diagram of a wind turbine variable speed pitch control system based on active disturbance rejection control according to an embodiment of the application. As shown in fig. 2, a variable speed and pitch control system 200 of a wind turbine based on active disturbance rejection control according to an embodiment of the present application includes: the data acquisition module 210 is configured to acquire wind speed values at a plurality of predetermined time points including a current time point, pitch angles at the plurality of predetermined time points including the current time point, and output power of the wind turbine generator at the plurality of predetermined time points including the current time point, which are acquired by a wind speed sensor; the wind speed data coding module 220 is configured to arrange the wind speed values at the multiple predetermined time points including the current time point as a wind speed input vector, and then obtain a multi-scale wind speed feature vector through the first multi-scale neighborhood feature extraction module; the pitch angle data encoding module 230 is configured to arrange the pitch angles of the multiple predetermined time points including the current time point as a pitch angle input vector, and then obtain a multi-scale pitch angle feature vector through the second multi-scale neighborhood feature extraction module; the power data encoding module 240 is configured to arrange the output power of the wind turbine generator at multiple predetermined time points including the current time point into a power input vector, and then obtain a multi-scale power feature vector through a third multi-scale neighborhood feature extraction module; a bayesian fusion module 250 for fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector using a bayesian probability model to obtain a posterior probability vector; a posterior information correction module 260, configured to correct the eigenvalues of each position in the posterior probability vector based on the mean and variance of the eigenvalue sets of all positions in the posterior probability vector to obtain a corrected posterior probability vector; and a control result generating module 270, configured to pass the corrected posterior probability vector through a classifier to obtain a classification result, where the classification result is used to indicate that the pitch angle at the current time point should be increased or decreased.

Specifically, in the embodiment of the present application, the data acquisition module 210 is configured to acquire wind speed values at a plurality of predetermined time points including a current time point, pitch angles at the plurality of predetermined time points including the current time point, and output power of the wind turbine at the plurality of predetermined time points including the current time point, which are acquired by a wind speed sensor. As mentioned above, it is considered that when the pitch system of the wind turbine is adjusted to control the wind power conversion efficiency to protect the wind turbine from damage, the existing pitch system is adjusted based on the wind conditions at the current point in time, resulting in an adjusted pitch angle that is not able to keep the output power and the rotor speed at their nominal values, which in turn would exacerbate the grid fluctuation. Therefore, in the technical scheme of the application, data prediction is expected to be dynamically carried out based on the wind speed values and the pitch angles of a plurality of preset time points including the current time point and the dynamic change characteristic information of the output power of the wind driven generator, so that the speed and pitch control of the wind driven generator is completed, and a Bayesian model is also introduced in the process for predictive adjustment, so that the predicted result is more accurate, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, the wind power conversion efficiency is controlled, and the wind driven generator is protected from being damaged.

That is, specifically, in the technical solution of the present application, first, wind speed values at a plurality of predetermined time points including a current time point are collected by a wind speed sensor, and pitch angles at the plurality of predetermined time points including the current time point are collected by an angle sensor, and output power of the wind power generator at the plurality of predetermined time points including the current time point is collected by a power detector.

Specifically, in this embodiment of the application, the wind speed data encoding module 220 is configured to arrange the wind speed values at the plurality of predetermined time points including the current time point as a wind speed input vector, and then obtain a multi-scale wind speed feature vector through a first multi-scale neighborhood feature extraction module. That is, in the technical solution of the present application, a convolutional neural network having an excellent performance in local implicit relevance feature extraction is further used to perform implicit feature extraction of data feature relevance. It should be understood that convolutional neural networks were originally models applied in the image domain, but the idea of local feature extraction can be applied to time series data analysis as well. For example, a time-series convolution structure with a convolution kernel size of 3, the convolution kernel is moved along the time dimension in the form of a sliding window for time-series data input, and outputs a weighted sum of the data within each time-series segment. Each convolution unit stacks a plurality of convolution kernels to output a multi-dimensional feature. The large convolution kernel can extract features from a large-scale time sequence neighborhood, wherein the influence of each item of numerical value in the neighborhood is smaller, so that the fluctuation of input data is weakened, and the influence of noise points on output features is reduced. However, the difference of numerical value change is weakened by the large-scale convolution kernel, and the problem of smooth transition is easily caused, so that the output characteristic loses the discrimination capability. In contrast, small scale convolution kernels are better able to retain information in the input data, but are also more susceptible to interference from noise therein. Therefore, the convolution units with different sizes are combined to extract the features of different time sequence scales in consideration of the characteristics of convolution with different scales. And then completing feature fusion by adopting a feature splicing mode, thereby obtaining the multi-scale neighborhood features.

More specifically, in an embodiment of the present application, the wind speed data encoding module includes: firstly, arranging the wind speed values of a plurality of preset time points including the current time point as the wind speed input vector; then, inputting the wind speed input vector into a first convolution layer of the first multi-scale neighborhood characteristic extraction module to obtain a first neighborhood scale wind speed associated characteristic vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; then, inputting the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood region scale wind speed associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and finally, cascading the first neighborhood scale wind speed associated feature vector and the second neighborhood scale wind speed associated feature vector to obtain the multi-scale wind speed feature vector.

FIG. 3 illustrates a block diagram of a wind turbine speed and pitch data encoding module in the wind turbine speed and pitch control system based on active disturbance rejection control according to an embodiment of the application. As shown in fig. 3, the wind speed data encoding module 220 includes: a wind speed vectorization unit 221, configured to arrange the wind speed values at the plurality of predetermined time points including the current time point as the wind speed input vector; a wind speed first scale convolution coding unit 222, configured to input the wind speed input vector into a first convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale wind speed associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a wind speed second scale convolution coding unit 223, configured to input the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale wind speed associated feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and a wind speed multi-scale feature cascading unit 224, configured to cascade the first neighborhood scale wind speed associated feature vector and the second neighborhood scale wind speed associated feature vector to obtain the multi-scale wind speed feature vector.

Specifically, in this embodiment of the application, the pitch angle data encoding module 230 and the power data encoding module 240 are configured to arrange the pitch angles at the plurality of predetermined time points including the current time point as a pitch angle input vector, pass through a second multi-scale neighborhood feature extraction module to obtain a multi-scale pitch angle feature vector, and arrange the output powers of the wind turbine at the plurality of predetermined time points including the current time point as a power input vector, pass through a third multi-scale neighborhood feature extraction module to obtain a multi-scale power feature vector. That is, in the technical solution of the present application, similarly, regarding the pitch angles at the plurality of predetermined time points including the current time point and the output powers of the wind power generators at the plurality of predetermined time points including the current time point, considering that the pitch angles and the output powers of the force generators also have dynamic implicit change characteristics in the time dimension, in order to extract the multi-scale neighborhood correlation of the dynamic change characteristics, the pitch angles at the plurality of predetermined time points including the current time point and the output powers of the wind power generators at the plurality of predetermined time points including the current time point are further input into the multi-scale neighborhood characteristic extraction module and processed to obtain a multi-scale feature vector and a multi-scale power feature vector. Therefore, the output correlation characteristics in neighborhoods with different scales of the dynamic change characteristics of the pitch angle and the output power of the wind driven generator not only contain the characteristics after smoothing, but also store the original input characteristic information, so as to avoid the loss of the information and further improve the accuracy of subsequent classification.

More specifically, in an embodiment of the present application, the pitch angle data encoding module comprises: a pitch angle vectorization unit, configured to arrange the pitch angles of the multiple predetermined time points including the current time point as the pitch angle input vector; a pitch angle first scale convolution coding unit, configured to input the pitch angle input vector into a first convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale pitch angle associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length; a pitch angle second scale convolution encoding unit configured to input the pitch angle input vector into a second convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale pitch angle associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and the pitch angle multi-scale feature cascading unit is used for cascading the first neighborhood scale pitch angle associated feature vector and the second neighborhood scale pitch angle associated feature vector to obtain the multi-scale pitch angle feature vector.

More specifically, in this embodiment, the power data encoding module includes: the power vectorization unit is used for arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into the power input vector; a power first scale convolution coding unit, configured to input the power input vector into a first convolution layer of the third multi-scale neighborhood feature extraction module to obtain a first neighborhood scale power correlation feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a power second scale convolution coding unit, configured to input the power input vector into a second convolution layer of the third multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale power correlation feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the power multi-scale feature cascading unit is used for cascading the first neighborhood scale power association feature vector and the second neighborhood scale power association feature vector to obtain the multi-scale power feature vector.

Specifically, in the embodiment of the present application, the bayesian fusion module 250 is configured to fuse the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a bayesian probability model to obtain an a posteriori probability vector. It should be appreciated that in view of using the multi-scale pitch angle feature vector as a prior probability, the technical solution in the present application aims at updating the prior probability to get a posterior probability upon new evidence, i.e. upon a new change of the wind speed value. Then, according to a bayesian formula, a posterior probability is a prior probability multiplied by an event probability divided by an evidence probability, and therefore, in the technical solution of the present application, a bayesian probability model is used to fuse the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector to obtain a posterior probability vector, where the multi-scale pitch angle feature vector is used as a prior, the multi-scale wind speed feature vector is used as an event, and the multi-scale power feature vector is used as an evidence.

More specifically, in this embodiment of the present application, the bayesian fusion module is further configured to: fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain the posterior probability vector;

wherein the formula is:

qi＝pi*ai/bi

Specifically, in this embodiment of the present application, the posterior information correction module 260 is configured to correct the eigenvalue of each position in the posterior probability vector based on the mean and the variance of the eigenvalue set of all positions in the posterior probability vector to obtain a corrected posterior probability vector. It should be understood that, since the bayesian probability model performs point-by-point calculation on the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector, the obtained posterior probability vector may have a problem of poor expression consistency among feature values, thereby affecting the classification effect. Therefore, in the technical solution of the present application, the posterior probability vector is further subjected to recursive squeeze-excitation optimization in length dimension. Here, the length-dimension recursive squeeze-excitation optimization activates the dimension recursion along the length direction of the feature distribution based on the statistical characteristics of the feature set, so as to infer the distribution of each sampling point of the feature in the length dimension thereof, and a predetermined-dimension squeeze-excitation mechanism composed of a ReLU-Sigmoid function is adopted to obtain a predetermined-dimension attention-enhanced confidence value, so as to improve the expression consistency of the high-dimension feature flow pattern of the posterior probability vector in the length dimension, and further improve the classification accuracy.

More specifically, in this embodiment of the present application, the posterior information correction module is further configured to: based on the mean and variance of the feature value sets of all positions in the posterior probability vector, correcting the feature value of each position in the posterior probability vector by the following formula to obtain the corrected posterior probability vector;

wherein the formula is:

Specifically, in the embodiment of the present application, the control result generating module 270 is configured to pass the corrected posterior probability vector through a classifier to obtain a classification result, where the classification result is used to indicate that the pitch angle at the current time point should be increased or decreased. In one specific example, the corrected a posteriori probability vectors are processed using the classifier to obtain the classification result with the following formula: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) I X }, wherein W ₁ To W _n As a weight matrix, B ₁ To B _n Is a bias vector and X is the corrected posterior probability vector.

In summary, the active disturbance rejection control-based wind turbine speed and pitch control system 200 is illustrated according to the embodiment of the present application, and employs an artificial intelligence control technique, and extracts a multi-scale neighborhood correlation of wind speed values and pitch angles at a plurality of predetermined time points including a current time point and a dynamic change characteristic of output power of a wind turbine through a multi-scale neighborhood characteristic extraction module, so as to dynamically predict data, thereby completing speed and pitch control of the wind turbine, and a bayesian model is introduced in the process to perform predictive adjustment, thereby making a predicted result more accurate. Therefore, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, and the wind power conversion efficiency is controlled to protect the fan from being damaged.

As described above, the wind turbine variable speed and pitch control system 200 based on the active disturbance rejection control according to the embodiment of the present application may be implemented in various terminal devices, such as a server of a wind turbine variable speed and pitch control algorithm based on the active disturbance rejection control. In one example, the active disturbance rejection control-based wind turbine variable speed pitch control system 200 according to an embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the wind turbine variable speed and pitch control system 200 based on active disturbance rejection control may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the wind turbine variable speed and pitch control system 200 based on active disturbance rejection control may also be one of many hardware modules of the terminal device.

Alternatively, in another example, the active disturbance rejection control-based wind turbine variable speed pitch control system 200 and the terminal device may also be separate devices, and the active disturbance rejection control-based wind turbine variable speed pitch control system 200 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information according to an agreed data format.

Exemplary method

FIG. 4 illustrates a flow chart of a wind turbine variable speed and pitch control method based on active disturbance rejection control. As shown in fig. 4, a method for controlling variable speed and variable pitch of a wind turbine based on active disturbance rejection control according to an embodiment of the present application includes the steps of: s110, acquiring wind speed values of a plurality of preset time points including a current time point, pitch angles of the plurality of preset time points including the current time point and output power of the wind driven generator of the plurality of preset time points including the current time point, wherein the wind speed values are acquired by a wind speed sensor; s120, arranging the wind speed values of the plurality of preset time points including the current time point as a wind speed input vector, and then obtaining a multi-scale wind speed feature vector through a first multi-scale neighborhood feature extraction module; s130, arranging the pitch angles of the multiple preset time points including the current time point as a pitch angle input vector, and then obtaining a multi-scale pitch angle feature vector through a second multi-scale neighborhood feature extraction module; s140, arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into a power input vector, and then obtaining a multi-scale power feature vector through a third multi-scale neighborhood feature extraction module; s150, fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model to obtain a posterior probability vector; s160, correcting the characteristic values of all the positions in the posterior probability vector based on the mean value and the variance of the characteristic value sets of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and S170, passing the corrected posterior probability vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the pitch angle of the current time point should be increased or decreased.

Fig. 5 illustrates an architecture diagram of a wind turbine variable speed and pitch control method based on active disturbance rejection control according to an embodiment of the application. As shown in fig. 5, in the network architecture of the wind turbine variable speed and pitch control method based on active disturbance rejection control, firstly, after arranging the obtained wind speed values (e.g., P1 as illustrated in fig. 5) at a plurality of predetermined time points including the current time point as a wind speed input vector (e.g., V1 as illustrated in fig. 5), a multi-scale wind speed feature vector (e.g., VF1 as illustrated in fig. 5) is obtained through a first multi-scale neighborhood feature extraction module (e.g., MS1 as illustrated in fig. 5); then, arranging the obtained pitch angles (e.g., P2 as illustrated in fig. 5) of the plurality of predetermined time points including the current time point as a pitch angle input vector (e.g., V2 as illustrated in fig. 5) and then passing through a second multi-scale neighborhood feature extraction module (e.g., MS2 as illustrated in fig. 5) to obtain a multi-scale pitch angle feature vector (e.g., VF2 as illustrated in fig. 5); then, arranging the obtained output power (for example, P3 as illustrated in fig. 5) of the wind turbine at a plurality of predetermined time points including the current time point into a power input vector (for example, V3 as illustrated in fig. 5), and then passing through a third multi-scale neighborhood feature extraction module (for example, MS3 as illustrated in fig. 5) to obtain a multi-scale power feature vector (for example, VF3 as illustrated in fig. 5); then, fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector using a bayesian probabilistic model (e.g., BPM as illustrated in fig. 5) to obtain an a posteriori probability vector (e.g., VP1 as illustrated in fig. 5); then, based on the mean and variance of the feature value sets of all positions in the posterior probability vector, correcting the feature values of each position in the posterior probability vector to obtain a corrected posterior probability vector (for example, VP2 as illustrated in fig. 5); and, finally, passing the corrected a posteriori probability vector through a classifier (e.g. circle S as illustrated in fig. 5) to obtain a classification result indicating that the pitch angle at the current point in time should be increased or decreased.

More specifically, in step S110 and step S120, wind speed values of a plurality of predetermined time points including a current time point, which are collected by a wind speed sensor, pitch angles of the plurality of predetermined time points including the current time point, and output power of the wind turbine generator of the plurality of predetermined time points including the current time point are obtained, and after the wind speed values of the plurality of predetermined time points including the current time point are arranged as a wind speed input vector, a multi-scale wind speed feature vector is obtained through a first multi-scale neighborhood feature extraction module. It should be appreciated that, given that the pitch system of a wind turbine is regulated to control the wind power conversion efficiency to protect the wind turbine from damage, existing pitch systems are adjusted based on the current point in time wind conditions, resulting in an adjusted pitch angle that is unable to maintain the output power and rotor speed at their nominal values, which in turn can exacerbate grid fluctuations. Therefore, in the technical scheme of the application, data prediction is expected to be dynamically performed based on the wind speed values and the pitch angles of a plurality of preset time points including the current time point and the dynamic change characteristic information of the output power of the wind driven generator, so that the speed and pitch control of the wind driven generator is completed, and a Bayesian model is introduced in the process for predictive adjustment, so that the predicted result is more accurate, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, the wind power conversion efficiency is controlled, and the wind driven generator is protected from being damaged.

It will be appreciated, then, that convolutional neural networks were originally models applied in the image domain, but the idea of local feature extraction can be applied to time series data analysis as well. For example, a time-series convolution structure with a convolution kernel size of 3, the convolution kernel is moved along the time dimension in the form of a sliding window for time-series data input, and outputs a weighted sum of the data within each time-series segment. Each convolution unit stacks a plurality of convolution kernels to output a multi-dimensional feature. The large convolution kernel can extract features from a large-scale time sequence neighborhood, wherein the influence of each item of value in the neighborhood is smaller, so that the fluctuation of input data is weakened, and the influence of noise points on output features is relieved. However, the difference of numerical value changes is weakened by the large-scale convolution kernel, and the problem of excessive smoothness is easily caused, so that the output characteristics lose the discrimination capability. In contrast, small scale convolution kernels are better able to retain information in the input data, but are also more susceptible to interference from noise therein. Therefore, in consideration of the characteristics of convolution with different scales, convolution units with different sizes are combined to extract the characteristics of different time sequence scales. And then completing feature fusion by adopting a feature splicing mode, thereby obtaining the multi-scale neighborhood features.

More specifically, in step S130 and step S140, after the pitch angles of the plurality of predetermined time points including the current time point are arranged as pitch angle input vectors, the pitch angles are passed through a second multi-scale neighborhood feature extraction module to obtain multi-scale pitch angle feature vectors, and after the output powers of the wind turbine generators of the plurality of predetermined time points including the current time point are arranged as power input vectors, the output powers are passed through a third multi-scale neighborhood feature extraction module to obtain multi-scale power feature vectors. That is, in the technical solution of the present application, similarly, regarding the pitch angles at the plurality of predetermined time points including the current time point and the output powers of the wind power generators at the plurality of predetermined time points including the current time point, considering that the pitch angles and the output powers of the force generators also have dynamic implicit change characteristics in the time dimension, in order to extract the multi-scale neighborhood correlation of the dynamic change characteristics, the pitch angles at the plurality of predetermined time points including the current time point and the output powers of the wind power generators at the plurality of predetermined time points including the current time point are further input into the multi-scale neighborhood characteristic extraction module and processed to obtain a multi-scale feature vector and a multi-scale power feature vector. Therefore, the output pitch angle and the intra-neighborhood associated features of the dynamic change features of the output power of the wind driven generator with different scales not only contain the smoothed features, but also save the original input feature information, so that the information loss is avoided, and the accuracy of subsequent classification can be improved.

More specifically, in steps S150 and S160, a bayesian probability model is used to fuse the multi-scale pitch angle eigenvector, the multi-scale wind speed eigenvector and the multi-scale power eigenvector to obtain a posterior probability vector, and the eigenvalue of each position in the posterior probability vector is corrected based on the mean and variance of the eigenvalue sets of all positions in the posterior probability vector to obtain a corrected posterior probability vector. It should be understood that, considering the use of the multi-scale pitch angle feature vector as a prior probability, the technical solution in the present application aims to update the prior probability to get a posterior probability in case of new evidence, i.e. in case of a new change in the wind speed value. Then, according to a bayesian formula, a posterior probability is a prior probability multiplied by an event probability divided by an evidence probability, and therefore, in the technical solution of the present application, a bayesian probability model is used to fuse the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector to obtain a posterior probability vector, where the multi-scale pitch angle feature vector is used as a prior, the multi-scale wind speed feature vector is used as an event, and the multi-scale power feature vector is used as an evidence.

However, since the bayesian probability model calculates the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector point by point, the obtained posterior probability vector may have a problem of poor expression consistency among feature values, thereby affecting the classification effect. Therefore, in the technical solution of the present application, the posterior probability vector is further subjected to recursive squeeze-excitation optimization of length dimension. Here, the length-dimension recursive squeeze-excitation optimization activates the dimension recursion along the length direction of the feature distribution based on the statistical characteristics of the feature set, so as to infer the distribution of each sampling point of the feature in the length dimension thereof, and a predetermined-dimension squeeze-excitation mechanism composed of a ReLU-Sigmoid function is adopted to obtain a predetermined-dimension attention-enhanced confidence value, so as to improve the expression consistency of the high-dimensional feature flow pattern of the posterior probability vector in the length dimension, thereby improving the classification accuracy.

More specifically, in step S170, the corrected posterior probability vector is passed through a classifier to obtain a classification result, which is used to indicate that the pitch angle at the current time point should be increased or decreased.

In summary, the active disturbance rejection control-based wind turbine speed and pitch control method based on the embodiment of the present application is illustrated, and an artificial intelligence control technology is adopted, and a multi-scale neighborhood feature extraction module is used for respectively extracting multi-scale neighborhood associations of wind speed values and pitch angles at a plurality of predetermined time points including a current time point and dynamic change features of output power of a wind turbine so as to dynamically predict data, thereby completing speed and pitch control of the wind turbine, and a bayesian model is introduced in the process for predictive adjustment, so that a predicted result is more accurate. Therefore, the output power and the rotor speed can be kept at the rated values under the condition of stabilizing the power grid fluctuation, and the wind power conversion efficiency is controlled to protect the fan from being damaged.

The basic principles of the present application have been described above with reference to specific embodiments, but it should be noted that advantages, effects, etc. mentioned in the present application are only examples and are not limiting, and the advantages, effects, etc. must not be considered to be possessed by various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.

The block diagrams of devices, apparatuses, devices, systems referred to in this application are only used as illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by one skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably herein. As used herein, the words "or" and "refer to, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".

It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations should be considered as equivalents of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.

Claims

1. A wind driven generator variable speed and variable pitch control system based on active disturbance rejection control is characterized by comprising:

the posterior information correction module is used for correcting the characteristic values of all positions in the posterior probability vector based on the mean value and the variance of the characteristic value sets of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and

2. The active disturbance rejection control-based wind turbine variable speed and pitch control system according to claim 1, wherein the wind speed data encoding module comprises:

the wind speed vectorization unit is used for arranging the wind speed values of a plurality of preset time points including the current time point into the wind speed input vector;

a wind speed first scale convolution coding unit, configured to input the wind speed input vector into a first convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale wind speed associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length;

a wind speed second scale convolution coding unit, configured to input the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale wind speed associated feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and

and the wind speed multi-scale characteristic cascading unit is used for cascading the first neighborhood scale wind speed correlation characteristic vector and the second neighborhood scale wind speed correlation characteristic vector to obtain the multi-scale wind speed characteristic vector.

3. The active disturbance rejection control based wind turbine variable speed and pitch control system according to claim 2, wherein the pitch angle data encoding module comprises:

a pitch angle vectorization unit, configured to arrange the pitch angles of the multiple predetermined time points including the current time point as the pitch angle input vector;

a pitch angle first scale convolution coding unit, configured to input the pitch angle input vector into a first convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale pitch angle associated feature vector, where the first convolution layer has a first one-dimensional convolution kernel of a first length;

a pitch angle second scale convolution encoding unit, configured to input the pitch angle input vector into a second convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale pitch angle associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and

and the pitch angle multi-scale feature cascading unit is used for cascading the first neighborhood scale pitch angle correlation feature vector and the second neighborhood scale pitch angle correlation feature vector to obtain the multi-scale pitch angle feature vector.

4. The active disturbance rejection control-based wind turbine variable speed and pitch control system according to claim 3, wherein the power data encoding module comprises:

the power vectorization unit is used for arranging the output power of the wind driven generator at a plurality of preset time points including the current time point into the power input vector;

a power first scale convolution coding unit, configured to input the power input vector into a first convolution layer of the third multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale power correlation feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length;

a power second scale convolution coding unit, configured to input the power input vector into a second convolution layer of the third multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale power correlation feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and

a power multi-scale feature cascading unit, configured to cascade the first neighborhood scale power association feature vector and the second neighborhood scale power association feature vector to obtain the multi-scale power feature vector.

5. The active disturbance rejection control-based wind turbine variable speed and pitch control system according to claim 4, wherein the Bayesian fusion module is further configured to: fusing the multi-scale pitch angle feature vector, the multi-scale wind speed feature vector and the multi-scale power feature vector by using a Bayesian probability model according to the following formula to obtain the posterior probability vector;

wherein the formula is:

qi＝pi*ai/bi

6. The active disturbance rejection control-based wind turbine variable speed and pitch control system according to claim 5, wherein the posterior information correction module is further configured to: based on the mean and variance of the feature value sets of all positions in the posterior probability vector, correcting the feature values of all positions in the posterior probability vector by the following formula to obtain the corrected posterior probability vector;

wherein the formula is:

wherein v is _i Representing the posterior probability vectorRepresenting the feature value of each position in the corrected posterior probability vector, and μ and σ are feature sets v _i E.v, V represents the posterior probability vector, exp (-) represents the exponential operation of the vector, the exponential operation raised to the power of the vector represents the natural exponential function value raised to the power of the value at each position of the vector.

7. The active disturbance rejection control-based wind turbine variable speed and pitch control system according to claim 6, wherein the control result generation module is further configured to: processing the corrected a posteriori probability vector using the classifier to obtain the classification result with the formula: softmax { (W) _n ,B _n ):…:(W ₁ ,B ₁ ) I X }, wherein W ₁ To W _n As a weight matrix, B ₁ To B _n Is a bias vector and X is the corrected posterior probability vector.

8. A wind driven generator speed and pitch control method based on active disturbance rejection control is characterized by comprising the following steps:

after the wind speed values of the multiple preset time points including the current time point are arranged as a wind speed input vector, a multi-scale wind speed feature vector is obtained through a first multi-scale neighborhood feature extraction module;

after the pitch angles of the plurality of preset time points including the current time point are arranged into pitch angle input vectors, a multi-scale pitch angle feature vector is obtained through a second multi-scale neighborhood feature extraction module;

correcting the characteristic value of each position in the posterior probability vector based on the mean value and the variance of the characteristic value set of all the positions in the posterior probability vector to obtain a corrected posterior probability vector; and

9. The active disturbance rejection control-based variable speed and variable pitch control method of the wind turbine according to claim 8, wherein the step of arranging the wind speed values at the plurality of predetermined time points including the current time point as a wind speed input vector and then obtaining a multi-scale wind speed feature vector through a first multi-scale neighborhood feature extraction module comprises:

arranging the wind speed values of a plurality of preset time points including the current time point into the wind speed input vector;

inputting the wind speed input vector into a first convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a first neighborhood scale wind speed associated feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length;

inputting the wind speed input vector into a second convolution layer of the first multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale wind speed associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and

and cascading the first neighborhood scale wind speed associated feature vector and the second neighborhood scale wind speed associated feature vector to obtain the multi-scale wind speed feature vector.

10. The active disturbance rejection control-based variable speed and variable pitch control method for the wind turbine according to claim 9, wherein the step of arranging the pitch angles of the plurality of predetermined time points including the current time point as a pitch angle input vector and then obtaining a multi-scale pitch angle feature vector through a second multi-scale neighborhood feature extraction module comprises:

arranging the pitch angles of a plurality of preset time points including the current time point into the pitch angle input vector;

inputting the pitch angle input vector into a first convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a first neighborhood region scale pitch angle correlation feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length;

inputting the pitch angle input vector into a second convolution layer of the second multi-scale neighborhood region feature extraction module to obtain a second neighborhood scale pitch angle associated feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, the first length being different from the second length; and

cascading the first neighborhood scale pitch angle association feature vector and the second neighborhood scale pitch angle association feature vector to obtain the multi-scale pitch angle feature vector.