Wind turbine generator temperature control method and system based on data analysis

Abstract

The invention provides a wind turbine generator temperature control method and system based on data analysis, and relates to the field of wind turbine generator control, and the wind turbine generator temperature control method based on data analysis comprises the following steps: constructing time sequence data; analyzing a temperature transfer path between the generator and other equipment on the basis of a variable mode decomposition method fused with a Talbot energy operator; according to a temperature transmission path between the generator and other equipment, and according to a collaborative optimization algorithm, adopting a generator following mode to perform collaborative control on the temperature of the generator and other equipment, and generating a control strategy; according to the generated control strategy, operation parameters of the generator and other equipment are adjusted in real time, temperature responses of the generator and other equipment are continuously monitored, and a feedback mechanism is used for evaluating the control effect. According to the invention, the coordination and effectiveness of temperature control are ensured, so that the operation efficiency of the whole wind turbine generator is improved.

Description

Technical Field

[0001] The present invention relates to the field of wind turbine generator control, and in particular to a method and system for controlling the temperature of a wind turbine generator based on data analysis. Background Art

[0002] A wind turbine is a complex system, the core components of which include a wind rotor and a generator. The wind rotor, as a key part for capturing wind energy, is finely constructed with blades, a hub, and reinforcements. The blades rotate under the action of wind power, which in turn drives the generator head to rotate, realizing the conversion of wind energy into mechanical energy, and finally converting mechanical energy into electrical energy through the generator. In the entire power supply system of wind power generation, in addition to the wind turbine itself, it also includes several important components such as the tower supporting the generator set, the battery charge controller, the inverter, the unloader, the grid connection controller, and the battery pack. These components work together to ensure the stable and efficient operation of the wind power generation system.

[0003] However, as the core component of a wind turbine generator set, the temperature control problem of the generator has always been a technical difficulty. In the existing technology, linear correlation or simple statistical methods are often used to analyze the heat transfer relationship between the generator and other equipment. However, this method has obvious limitations. It cannot fully reveal the nonlinear intensity and directionality of temperature transfer, resulting in an insufficient understanding of the heat transfer path. At the same time, traditional thermodynamic models are also unable to describe the complex temperature dynamic relationship between the generator and other equipment in a wind turbine. These models are often oversimplified and cannot accurately depict the thermal coupling effect between equipment, which affects the formulation and implementation of temperature control strategies.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of this, the present invention provides a wind turbine generator temperature control method and system based on data analysis to solve the above-mentioned problems.

[0006] In order to solve the above problems, the specific technical solutions adopted by the present invention are as follows:

[0007] According to one aspect of the present invention, a method for controlling the temperature of a wind turbine generator based on data analysis is provided. The method for controlling the temperature of a wind turbine generator based on data analysis comprises the following steps:

[0008] S1. Obtain the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtain the environmental interference evaluation value based on the comprehensive analysis of the environmental interference data, obtain the temperature data correction value based on the matching of the environmental interference evaluation value, and construct the time series data based on the timestamp information of the temperature data and the temperature data correction value; analyze the temperature transfer path between the generator and other equipment based on the variable mode decomposition method fused with the Tiger energy operator and combined with the transfer entropy theory;

[0009] S2. Based on the temperature transfer path between the generator and other equipment and the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner to generate a control strategy;

[0010] S3. According to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored to obtain the sensor response data. The sensor response index is obtained by comprehensive analysis based on the sensor response data. The transmission quality evaluation value is obtained by comprehensive analysis based on the sensor response index and the environmental interference evaluation value. The control effect is evaluated based on the transmission quality evaluation value and the feedback mechanism.

[0011] Preferably, the temperature data of the generator and other equipment in the wind turbine are obtained, and time series data is constructed according to the timestamp information of the temperature data; the variable mode decomposition method based on the fusion of the Tiger energy operator and the transfer entropy theory are combined to analyze the temperature transfer path between the generator and other equipment, including the following steps:

[0012] S11, respectively collecting temperature data of the generator and other equipment in the wind turbine, and constructing time series data based on the timestamp information of the temperature data;

[0013] S12, the environmental interference data, including wind speed, electromagnetic interference intensity and ambient temperature, obtains the critical wind speed, critical electromagnetic interference intensity, reference ambient temperature and allowable deviation ambient temperature from the temperature control database, comprehensively analyzes to obtain the environmental interference evaluation value, and obtains the temperature data correction value according to the matching of the environmental interference evaluation value;

[0014] S13, using the temperature data correction value to correct the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data by the variable mode decomposition method to obtain the mode function, and optimizing each mode function by using the Tiger energy operator and singular value decomposition to obtain the reconstructed and denoised time series data;

[0015] S14. Calculate the temperature transfer entropy between the generator and other devices by using the reconstructed and denoised time series data, and analyze the temperature transfer path between the generator and other devices based on the calculation result of the temperature transfer entropy.

[0016] Preferably, obtaining the temperature data correction value according to the environmental interference evaluation value matching comprises the following steps:

[0017] The environmental interference evaluation value is matched with the temperature data correction value corresponding to each environmental interference evaluation value preset in the temperature control database to obtain the temperature data correction value.

[0018] Preferably, the temperature data in the time series data is corrected by using the temperature data correction value to obtain the corrected temperature data, the corrected temperature data is decomposed by the variable mode decomposition method to obtain the mode function, and each mode function is optimized by using the Tiger energy operator and singular value decomposition to obtain the reconstructed denoised time series data, which includes the following steps:

[0019] S131, adding the temperature data correction value to the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data into multiple intrinsic mode function components by using the variable mode decomposition method, and screening the multiple intrinsic mode function components by calculating the cross-correlation and the mean square error;

[0020] S132, using the Tiger energy operator to process the eigenmode function components after the screening process, by extracting the instantaneous amplitude and calculating the instantaneous power of the eigenmode function components, to obtain the instantaneous power sequence of the eigenmode function components;

[0021] S133. Based on the instantaneous power sequence of the intrinsic mode function components, singular value decomposition is performed on each intrinsic mode function component, and each intrinsic mode function component processed by the singular value decomposition is reconstructed to obtain the reconstructed and denoised time series data.

[0022] Preferably, the temperature data correction value is added to the temperature data in the time series data to obtain the corrected temperature data, the corrected temperature data is decomposed into a plurality of intrinsic mode function components by the variable mode decomposition method, and the plurality of intrinsic mode function components are screened by calculating the cross-correlation and the mean square error, comprising the following steps:

[0023] S1311. Collect historical temperature data of generators and other equipment in the wind turbine generator set, and calculate standard temperature data of generators and other equipment in the wind turbine generator set by statistics;

[0024] S1312, adding the standard temperature data and the temperature data correction value to obtain the corrected temperature data, cross-correlating each intrinsic mode function component with the corrected temperature data, and calculating the cross-correlation value and mean square error value of the intrinsic mode function component and the corrected temperature data;

[0025] S1313, combining the cross-correlation value and the mean square error value into an objective function, and comparing the value of the objective function with a preset threshold, and performing screening processing on the intrinsic mode function components corresponding to the objective function that meets the preset threshold.

[0026] Preferably, the method of calculating the temperature transfer entropy between the generator and other devices by reconstructing the denoised time series data, and analyzing the temperature transfer path between the generator and other devices according to the calculation result of the temperature transfer entropy comprises the following steps:

[0027] S141, configuring a delay time according to the time series data after the generator and other devices are reconstructed and de-noised, and calculating the autocorrelation function of the time series data after the generator and other devices are reconstructed and de-noised;

[0028] S142, using the generator and other devices to reconstruct the autocorrelation function of the time series data after noise reduction, and calculating the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of the generator;

[0029] S143, calculating the temperature transfer entropy between the generator and the other equipment using the transfer entropy theory according to the linear correlation function of the temperature data between the generator and the other equipment and the normalized autocorrelation function of the generator and in combination with the delay time;

[0030] S144. Identify the temperature transfer path between the generator and other devices based on the calculated temperature transfer entropy.

[0031] Preferably, the autocorrelation function of the time series data after noise reduction is reconstructed by using the generator and other devices, the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of other devices are calculated respectively:

[0032]

[0033] In the formula, represents the normalized autocorrelation function of generator x;

[0034] τ represents the delay time;

[0035] R xx represents the autocorrelation function of generator x;

[0036] ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y;

[0037] R xy represents the cross-correlation function between the temperature data of the generator x and other equipment y;

[0038] R yyRepresents the autocorrelation function of other devices y.

[0039] Preferably, according to the linear correlation function of the temperature data between the generator and other equipment and the normalized autocorrelation function of the generator, and in combination with the delay time, the calculation formula for calculating the temperature transfer entropy between the generator and other equipment using the transfer entropy theory is:

[0040]

[0041] Where, T (y→x) It represents the temperature transfer entropy value of the temperature transferred from other device y to generator x;

[0042] k and l represent the embedding dimensions of the time series data of generator x and other equipment y, respectively;

[0043] τ represents the delay time;

[0044] ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y;

[0045] represents the normalized autocorrelation function of the generator x.

[0046] Preferably, according to the temperature transfer path between the generator and other equipment and according to the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner, and the control strategy is generated, including the following steps:

[0047] S21, taking the generator and other devices as nodes, and establishing a temperature transfer path model according to the temperature transfer path between the generator and other devices;

[0048] S22, constructing a target optimization function with the generator temperature as the core control target, and setting corresponding constraints;

[0049] S23. According to the target optimization function and the temperature transfer path model, the temperature adjustment method of the generator and other equipment is determined through a collaborative optimization algorithm, and a control strategy is generated according to the temperature adjustment method of the generator and other equipment.

[0050] Preferably, according to the target optimization function and the temperature transfer path model, determining the temperature adjustment method of the generator and other equipment through a collaborative optimization algorithm, and generating a control strategy according to the temperature adjustment method of the generator and other equipment includes the following steps:

[0051] S231, initializing the temperature adjustment method of the generator and other equipment based on the target optimization function and the temperature transfer path model;

[0052] S232, using a collaborative optimization algorithm to evaluate the performance of the current temperature adjustment method by calculating a fitness value, and generating a new temperature adjustment method through selection, crossover and mutation operations according to the evaluation result;

[0053] S233, repeating the evaluation, selection, crossover and mutation operations, gradually optimizing the temperature adjustment method, until the maximum number of iterations is reached, and obtaining the final temperature adjustment method;

[0054] S234. Determine the control strategy of the generator and other equipment according to the final temperature adjustment method.

[0055] Preferably, according to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored, the sensor response data is obtained, the sensor response index is obtained according to the comprehensive analysis of the sensor response data, the transmission quality evaluation value is obtained according to the comprehensive analysis of the sensor response index and the environmental interference evaluation value, and the control effect is evaluated according to the transmission quality evaluation value and the feedback mechanism, including the following steps:

[0056] S31. According to the generated control strategy, adjust the operating parameters of the generator and other equipment in real time, and continuously monitor the temperature response of the generator and other equipment;

[0057] S32, the sensor response data, including response time and data transmission speed, obtains the critical response time, reference data transmission speed and allowable deviation data transmission speed from the temperature control database, and comprehensively analyzes to obtain the sensor response index;

[0058] S33. Obtain a critical sensor response index and a critical environmental interference assessment value from a temperature control database, obtain a transmission quality assessment value based on a comprehensive analysis of the sensor response index, the environmental interference assessment value, the critical sensor response index and the critical environmental interference assessment value, perform evaluation feedback based on the transmission quality assessment value, and use a feedback mechanism to evaluate the control effect.

[0059] Preferably, performing evaluation feedback according to the transmission quality evaluation value comprises the following steps:

[0060] S331. Obtain the transmission quality assessment threshold from the temperature control database, compare the transmission quality assessment value with the transmission quality assessment threshold, if the transmission quality assessment value is less than the transmission quality assessment threshold, notify relevant personnel to perform maintenance, if the transmission quality assessment value is greater than or equal to the transmission quality assessment threshold, no additional processing is performed.

[0061] According to another aspect of the present invention, a wind turbine generator temperature control system based on data analysis is provided, the wind turbine generator temperature control system based on data analysis comprises: a temperature transfer path analysis module, a control strategy generation module and a control effect evaluation module;

[0062] The temperature transfer path analysis module is used to obtain the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtain the environmental interference evaluation value based on the comprehensive analysis of the environmental interference data, obtain the temperature data correction value based on the matching of the environmental interference evaluation value, and construct the time series data based on the timestamp information of the temperature data and the temperature data correction value; based on the variable mode decomposition method fused with the Tiger energy operator, and combined with the transfer entropy theory, the temperature transfer path between the generator and other equipment is analyzed;

[0063] A control strategy generation module is used to generate a control strategy by collaboratively controlling the temperature of the generator and other equipment in a generator following manner according to the temperature transfer path between the generator and other equipment and in accordance with a collaborative optimization algorithm;

[0064] The control effect evaluation module is used to adjust the operating parameters of the generator and other equipment in real time according to the generated control strategy, and continuously monitor the temperature response of the generator and other equipment, obtain sensor response data, and obtain the sensor response index based on the comprehensive analysis of the sensor response data. The sensor response index and the environmental interference evaluation value are comprehensively analyzed to obtain the transmission quality evaluation value, and the control effect is evaluated based on the transmission quality evaluation value and the feedback mechanism.

[0065] The beneficial effects of the present invention are:

[0066] 1. The present invention obtains the temperature data of the generator and other equipment in the wind turbine and constructs time series data. The method can monitor the temperature changes of each device in real time and accurately. The variable mode decomposition method integrating the Tiger energy operator and combining the transfer entropy theory can deeply analyze the temperature transfer path between the generator and other equipment, which is helpful to understand the internal mechanism of temperature change and provide a scientific basis for subsequent temperature control. Through the collaborative optimization algorithm, the method can generate a collaborative control strategy for the generator to follow the temperature changes of other equipment. This strategy takes into account the mutual influence between the various devices, ensures the coordination and effectiveness of temperature control, and thus improves the operating efficiency of the entire wind turbine.

[0067] 2. The present invention decomposes the temperature time series data into multiple intrinsic mode function components through the variable mode decomposition method, and optimizes the processing by combining the Tiger energy operator and singular value decomposition, which effectively removes the noise and redundant information in the original data, improves the accuracy and reliability of the data, and utilizes the reconstructed denoised time series data and combines the transfer entropy theory to calculate the temperature transfer entropy between the generator and other equipment, which can accurately identify the temperature transfer path between the various devices. By accurately calculating the temperature transfer entropy, the intensity and direction of the temperature transfer between the various devices can be quantified, thereby formulating a more targeted and effective temperature control strategy.

[0068] 3. The present invention uses generators and other equipment as nodes and establishes a model based on the temperature transfer path. This method realizes comprehensive and systematic control of the temperature of each device in the wind turbine. This systematic control method helps to ensure temperature coordination between the devices and avoid local overheating or insufficient cooling. It constructs a target optimization function with the generator temperature as the core control target and sets constraints to make the formulation of the control strategy more targeted and clear, which helps to ensure that the temperature of the generator can be preferentially and effectively controlled during the control process, thereby ensuring the safe operation of the core components of the wind turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0070] Figure 1 is a flow chart of a method for controlling temperature of a wind turbine generator based on data analysis according to an embodiment of the present invention;

[0071] Figure 2 The invention is a principle block diagram of a temperature control system of a wind turbine generator based on data analysis according to an embodiment of the invention.

[0072] In the figure:

[0073] 1. Temperature transfer path analysis module; 2. Control strategy generation module; 3. Control effect evaluation module. DETAILED DESCRIPTION

[0074] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0075] According to an embodiment of the present invention, a method and system for controlling the temperature of a wind turbine generator based on data analysis are provided.

[0076] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a method for controlling the temperature of a wind turbine generator based on data analysis is provided. The method for controlling the temperature of a wind turbine generator based on data analysis comprises the following steps:

[0077] S1. Obtain the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtain the environmental interference evaluation value based on the comprehensive analysis of the environmental interference data, obtain the temperature data correction value based on the matching of the environmental interference evaluation value, and construct the time series data based on the timestamp information of the temperature data and the temperature data correction value; analyze the temperature transfer path between the generator and other equipment based on the variable mode decomposition method fused with the Tiger energy operator and combined with the transfer entropy theory;

[0078] As a preferred implementation, the temperature data of the generator and other equipment in the wind turbine are obtained, and time series data is constructed according to the timestamp information of the temperature data; the variable mode decomposition method based on the fusion of the Tiger energy operator and the transfer entropy theory are used to analyze the temperature transfer path between the generator and other equipment, including the following steps:

[0079] S11, respectively collecting temperature data of the generator and other equipment in the wind turbine, and constructing time series data based on the timestamp information of the temperature data;

[0080] It should be noted that representative monitoring points are selected on the generator, such as bearings, windings, cooling systems and other key parts; at the same time, other key equipment in the wind turbine, such as gearboxes, pitch systems, yaw systems, etc., are selected and temperature monitoring points are also set on them.

[0081] By installing high-precision temperature sensors at each monitoring point, we ensure that the temperature changes of the equipment can be accurately measured and recorded.

[0082] When collecting temperature data, record the timestamp information corresponding to each piece of data. The timestamp should be accurate to seconds or finer time units. The collected temperature data is sorted in the order of timestamps and processed by the temperature data correction value to form time series data. For each monitoring point, an independent time series data set can be obtained.

[0083] S12, the environmental interference data, including wind speed, electromagnetic interference intensity and ambient temperature, obtains the critical wind speed, critical electromagnetic interference intensity, reference ambient temperature and allowable deviation ambient temperature from the temperature control database, comprehensively analyzes to obtain the environmental interference evaluation value, and obtains the temperature data correction value according to the matching of the environmental interference evaluation value;

[0084] The environmental interference assessment value is obtained as follows:

[0085]

[0086] In the formula, ξE1 represents the environmental interference assessment value, α1 represents the environmental interference assessment impact factor corresponding to the wind speed, α2 represents the environmental interference assessment impact factor corresponding to the electromagnetic interference intensity, α3 represents the environmental interference assessment impact factor corresponding to the ambient temperature, W1 represents the wind speed, W0 represents the critical wind speed, E1 represents the electromagnetic interference intensity, E0 represents the critical electromagnetic interference intensity, T1 represents the ambient temperature, T0 represents the reference ambient temperature, T2 represents the allowable deviation ambient temperature, and e is a natural constant.

[0087] α1, α2 and α3 are the environmental interference assessment impact factors corresponding to the wind speed, electromagnetic interference intensity and ambient temperature preset in the temperature control database, respectively. These impact factors are numerical indicators that measure the impact of the above interference parameters on the environmental interference assessment value. Specifically, there is a mapping relationship table for each of the wind speed, electromagnetic interference intensity and ambient temperature, which records each possible parameter value and its corresponding environmental interference assessment impact factor. These mapping relationships can be one-to-one or many-to-one. For example, in actual applications, when it is necessary to evaluate a certain environmental interference assessment value, the measured wind speed, electromagnetic interference intensity and ambient temperature can be input into their respective corresponding mapping relationship tables, and the environmental interference assessment impact factors corresponding to these values ​​can be quickly found, where the value range of the impact factor is between 0 and 1.

[0088] As a preferred implementation, obtaining the temperature data correction value according to the environmental interference evaluation value matching includes the following steps:

[0089] The environmental interference evaluation value is matched with the temperature data correction value corresponding to each environmental interference evaluation value preset in the temperature control database to obtain the temperature data correction value.

[0090] It should be noted that the environmental interference assessment value and the temperature data correction value corresponding to each environmental interference assessment value preset in the temperature control database form a mapping set, and the temperature data correction value corresponding to the environmental interference assessment value can be obtained by inputting the mapping set according to the environmental interference assessment value.

[0091] S13, using the temperature data correction value to correct the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data by the variable mode decomposition method to obtain the mode function, and optimizing each mode function by using the Tiger energy operator and singular value decomposition to obtain the reconstructed and denoised time series data;

[0092] As a preferred implementation, the temperature data in the time series data is corrected by using the temperature data correction value to obtain the corrected temperature data, the corrected temperature data is decomposed by the variable mode decomposition method to obtain the mode function, and each mode function is optimized by using the Tiger energy operator and singular value decomposition to obtain the reconstructed denoised time series data, including the following steps:

[0093] S131, adding the temperature data correction value to the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data into multiple intrinsic mode function components by using the variable mode decomposition method, and screening the multiple intrinsic mode function components by calculating the cross-correlation and the mean square error;

[0094] As a preferred implementation, the temperature data correction value is added to the temperature data in the time series data to obtain the corrected temperature data, the corrected temperature data is decomposed into multiple intrinsic mode function components by the variable mode decomposition method, and the multiple intrinsic mode function components are screened by calculating the cross-correlation and the mean square error, including the following steps:

[0095] S1311. Collect historical temperature data of generators and other equipment in the wind turbine generator set, and calculate standard temperature data of generators and other equipment in the wind turbine generator set by statistics;

[0096] It should be noted that the historical temperature data of the generator and other equipment are collected from the monitoring system of the wind turbine. It is ensured that the collected data contains a sufficient time span to reflect the temperature changes under different working conditions. The processed data is calculated using statistical methods to calculate the average value of all data points, and the standard temperature data of the generator and other equipment is determined based on the average value of all data points.

[0097] S1312, adding the standard temperature data and the temperature data correction value to obtain the corrected temperature data, cross-correlating each intrinsic mode function component with the corrected temperature data, and calculating the cross-correlation value and mean square error value between the intrinsic mode function component and the standard temperature data;

[0098] It should be noted that the corrected temperature data is based on the timestamp of the collected temperature data, and the correction value of the temperature data is added to the corresponding timestamp. The corrected temperature data is obtained as follows:

[0099] T corrected (t) = T(t) + C(t);

[0100] Where, T corrected (t) represents the corrected temperature data of the t-th temperature data timestamp, T(t) represents the temperature data of the t-th temperature data timestamp, C(t) represents the temperature data correction value, where t is the timestamp number of each temperature data, t=1, 2, 3, ..., M, and M is the total number of temperature data timestamps.

[0101] Specifically, the modified temperature data is decomposed into multiple IMF (multiple intrinsic mode function) components and a residual term (if any) by applying the variable mode decomposition algorithm to the corrected temperature data.

[0102] Among them, IMF component characteristics: Each IMF component should meet two conditions: first, in the entire data sequence, the number of extreme points and zero-crossing points must be equal or differ by at most one; second, at any point, the average value of the upper envelope defined by the local maximum point and the lower envelope defined by the local minimum point is zero.

[0103] In addition, cross-correlation is a method of measuring the similarity between two signals, which takes into account the time delay between the signals. For two signals x(t) and y(t), their cross-correlation function represents the degree of similarity between x(t) and y(t) after a time shift of τ. For each IMF component, the cross-correlation function between it and the corrected temperature data is calculated, and the delay time that makes the cross-correlation function reach the maximum value is found. Then, this maximum value is used as the cross-correlation value of the IMF component and the corrected temperature data.

[0104] Mean Squared Error (MSE) is a measure of the difference between two signals. For two signals x(t) and y(t), their mean square error is defined as E[(x(t)-y(t)) 2 ], where E represents the expectation operation. For each IMF component, the mean square error between it and the corrected temperature data is calculated, which can be obtained by calculating the square of the difference between the two signals and then finding the average value.

[0105] S1313, combining the cross-correlation value and the mean square error value into an objective function, and comparing the value of the objective function with a preset threshold, and performing screening processing on the intrinsic mode function components corresponding to the objective function that meets the preset threshold.

[0106] It should be noted that the cross-correlation value and mean square error value of each intrinsic mode function component are normalized. An objective function is constructed to comprehensively evaluate the importance of each modal component. The objective function expression is:

[0107] F k =α·ρ norm,k -β·MSE norm,k

[0108] In the formula, F k represents the objective function composed of the cross-correlation value and the mean square error value, α and β represent weight coefficients, which are used to balance the importance of cross-correlation and mean square error, ρ norm,k Represents the cross-correlation value of each intrinsic mode function component, MSE norm,k Represents the mean square error value of each intrinsic mode function component.

[0109] The threshold of the objective function is set based on experience or data analysis results. The threshold should be able to distinguish important modal components from noise or irrelevant components. If the threshold is not ideal, the threshold can be adjusted according to the screening results until a satisfactory result is obtained.

[0110] If the objective function is greater than the threshold, the modal component is retained; otherwise, it is regarded as noise or irrelevant component and removed.

[0111] S132, using the Tiger energy operator to process the eigenmode function components after the screening process, by extracting the instantaneous amplitude and calculating the instantaneous power of the eigenmode function components, to obtain the instantaneous power sequence of the eigenmode function components;

[0112] It should be noted that using the Teager Energy Operator (TEO) to process the intrinsic mode function (IMF) components after screening is an effective signal analysis method, which is particularly suitable for extracting the instantaneous characteristics of the signal. The Teager Energy Operator is a nonlinear operator that can directly act on the signal to extract the instantaneous energy or intensity of the signal. For discrete signals, the Teager Energy Operator is usually approximated by calculating the difference between adjacent sample points.

[0113] Specifically, the Tiger Energy operator is applied to each IMF component after the screening process, which usually involves calculating the first and second order differences of the signal and using these differences to estimate the instantaneous amplitude of the signal. The instantaneous amplitude can be understood as the "strength" or "size" of the signal at a certain moment. For an IMF component, it reflects the oscillation amplitude of the component at a specific moment.

[0114] Instantaneous power is the square of the instantaneous amplitude of the signal, which represents the power or energy density of the signal at a certain moment. For each IMF component, the instantaneous power of the component can be obtained by squaring its instantaneous amplitude. This will produce a new sequence with the same length as the IMF component, namely the instantaneous power sequence.

[0115] S133. Based on the instantaneous power sequence of the intrinsic mode function components, singular value decomposition is performed on each intrinsic mode function component, and each intrinsic mode function component processed by the singular value decomposition is reconstructed to obtain the reconstructed and denoised time series data.

[0116] It should be noted that the instantaneous power sequence of each IMF component is constructed into a matrix by segmenting the instantaneous power sequence and treating each segment as a row or column of a matrix. The size of the matrix (number of rows and columns) will depend on the segmentation method and the length of the IMF component.

[0117] Singular value decomposition is a matrix decomposition method that can decompose a matrix into the product of three matrices, namely UΣV^T, where U and V are orthogonal matrices, Σ is a diagonal matrix, and the elements on its diagonal are called singular values. In the result of singular value decomposition, the singular values ​​represent the distribution of energy or information in the matrix. Generally, larger singular values ​​correspond to the main information or pattern in the matrix, while smaller singular values ​​may represent noise or irrelevant information.

[0118] Based on the size and distribution of the singular values, a set of "important" singular values ​​are selected for reconstruction. This can be achieved by setting a threshold and only retaining singular values ​​larger than the threshold.

[0119] Using the selected singular values ​​and the corresponding U and V matrix parts, a denoised matrix that approximates the original matrix is ​​reconstructed. This process is called matrix reconstruction.

[0120] Convert each row or column of the denoised matrix (depending on how the original matrix was constructed) back to time series data. This will result in a denoised time series of the same length as the original IMF components.

[0121] S14. Calculate the temperature transfer entropy between the generator and other devices by using the reconstructed and denoised time series data, and analyze the temperature transfer path between the generator and other devices based on the calculation result of the temperature transfer entropy.

[0122] As a preferred implementation, the temperature transfer entropy between the generator and other devices is calculated by reconstructing the denoised time series data, and according to the calculation result of the temperature transfer entropy, the temperature transfer path between the generator and other devices is analyzed, which includes the following steps:

[0123] S141, configuring a delay time according to the time series data after the generator and other devices are reconstructed and de-noised, and calculating the autocorrelation function of the time series data after the generator and other devices are reconstructed and de-noised;

[0124] It should be noted that the delay time is set according to the sampling frequency of the time series data and the characteristics of the equipment, and the autocorrelation function R is calculated for the time series data after the generator is reconstructed and de-noised. xx (τ);

[0125]

[0126] In the formula, R xx (τ) represents the autocorrelation function of the time series data after generator reconstruction and denoising;

[0127] N represents the time series length of the time series data after the generator reconstruction and denoising;

[0128] t represents time;

[0129] x(t) represents the reconstructed and denoised time series data of the generator at time t;

[0130] x(t+τ) represents the reconstructed and denoised time series data of the generator at time t with a delay of τ;

[0131] The autocorrelation function of the time series data after reconstruction and noise reduction of other equipment can be calculated by using the calculation formula for calculating the autocorrelation function of the time series data after reconstruction and noise reduction of the generator.

[0132] S142, using the generator and other devices to reconstruct the autocorrelation function of the time series data after noise reduction, and calculating the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of the generator;

[0133] As a preferred implementation, the autocorrelation function of the time series data after noise reduction is reconstructed using the generator and other devices, and the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of other devices are calculated by the following formulas:

[0134]

[0135] R xy (τ)=E[x(t)y(t+τ)]

[0136] In the formula, represents the normalized autocorrelation function of generator x;

[0137] τ represents the delay time;

[0138] R xxrepresents the autocorrelation function of generator x;

[0139] ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y;

[0140] R xy represents the cross-correlation function between the temperature data of the generator x and other equipment y;

[0141] R yy represents the autocorrelation function of other devices y;

[0142] E stands for expectation;

[0143] x(t) represents the reconstructed and denoised time series data of generator x at time t;

[0144] y(t+τ) represents the reconstructed and denoised time series data of other device y at time t with a delay of τ.

[0145] S143, calculating the temperature transfer entropy between the generator and the other equipment using the transfer entropy theory according to the linear correlation function of the temperature data between the generator and the other equipment and the normalized autocorrelation function of the generator and in combination with the delay time;

[0146] As a preferred implementation, based on the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of the generator, and combined with the delay time, the temperature transfer entropy between the generator and other devices is calculated using the transfer entropy theory. The calculation formula is:

[0147]

[0148] Where, T (y→x) It represents the temperature transfer entropy value of the temperature transferred from other device y to generator x;

[0149] k and l represent the embedding dimensions of the time series data of generator x and other equipment y, respectively;

[0150] τ represents the delay time;

[0151] ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y;

[0152] represents the normalized autocorrelation function of the generator x.

[0153] S144. Identify the temperature transfer path between the generator and other devices based on the calculated temperature transfer entropy.

[0154] Specifically, through temperature transfer entropy, the temperature transfer relationship between the generator and other equipment can be analyzed, and the impact path and information flow can be clearly identified. The following steps are included:

[0155] The transfer entropy values ​​of all devices are organized into a matrix. Each entropy value represents the intensity of temperature transfer from the generator to a certain device. If the entropy value is zero or close to zero, it means that the temperature transfer from the generator to the device is not significant.

[0156] If the entropy value transferred from the generator to the device is T (x→y) Greater than the reverse transfer entropy value T from the device to the generator (y→x) , indicating that the temperature is transferred from the generator to the equipment. On the contrary, if T (y→x) >T (x→y) , indicating that the equipment has a greater reverse impact on the generator.

[0157] By comparing the entropy values, several paths with the largest temperature transfer intensity are screened out as the main temperature transfer paths.

[0158] For example, the transfer entropy value of the generator to device A and device B is significantly higher than that of other devices, indicating that these two paths are key temperature transfer paths.

[0159] S2. Based on the temperature transfer path between the generator and other equipment and the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner to generate a control strategy;

[0160] As a preferred implementation, according to the temperature transfer path between the generator and other equipment and according to the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner, and the control strategy is generated, including the following steps:

[0161] S21, taking the generator and other devices as nodes, and establishing a temperature transfer path model according to the temperature transfer path between the generator and other devices;

[0162] Specifically, the generators and other devices are regarded as nodes in the network, and a graph structure model is established based on the temperature transfer relationship between them.

[0163] Node: represents generators and other equipment.

[0164] Edge: represents the temperature transfer path between the generator and the equipment.

[0165] Edge weight: represents the intensity of temperature transfer (such as calculated by temperature transfer entropy).

[0166] For example, suppose a wind turbine consists of the following equipment:

[0167] Generator (Node G), gearbox (Node GBox), cooling system (Node C), pitch motor (Node P) and yaw system (Node Y)

[0168] The following temperature transfer paths exist between devices:

[0169] There is direct heat transfer between the generator and the gearbox.

[0170] The generator transfers heat indirectly to other equipment through the cooling system.

[0171] The pitch system and yaw system have a weak thermal connection to the cooling system.

[0172] The graph structure model is:

[0173] Node set: {G,GBox,C,P,Y}

[0174] G: generator, GBox: gearbox, C: cooling system, P: pitch motor, YY: yaw system.

[0175] Edge Set:

[0176] (G,GBox): Direct heat transfer between the generator and the gearbox.

[0177] (G,C): Heat transfer between generator and cooling system.

[0178] (C,P): Heat transfer between cooling system and pitch motor.

[0179] (C,Y): Heat transfer between cooling system and yaw system.

[0180] The graph structure of the network can be expressed as:

[0181] Node set: V = {G, GBox, C, P, Y};

[0182] Edge set: E = {(G,GBox),(G,C),(C,P),(C,Y)}.

[0183] S22, constructing a target optimization function with the generator temperature as the core control target, and setting corresponding constraints;

[0184] Specifically, the objective optimization function aims to keep the generator temperature within the ideal range by collaboratively adjusting the temperature states of the generator and other equipment, while taking into account the temperature stability of other equipment and the overall efficiency of the system.

[0185] Taking the generator temperature as the core control target, the target optimization function is in the form of:

[0186]

[0187] Where minF(H x ,H y ) represents minimizing the degree to which the generator temperature deviates from the target value and the temperatures of other equipment deviate from their respective target values;

[0188] H x Indicates the temperature of the generator;

[0189] H set Indicates the target temperature of the generator;

[0190] H y Indicates the temperature of other equipment;

[0191] H yi represents the temperature of the i-th device;

[0192] H set,i represents the target temperature of the i-th device;

[0193] r represents the weight coefficient of the generator temperature, reflecting its priority;

[0194] e i The weight coefficient representing the temperature of the i-th device.

[0195] S23. According to the target optimization function and the temperature transfer path model, the temperature adjustment method of the generator and other equipment is determined through a collaborative optimization algorithm, and a control strategy is generated according to the temperature adjustment method of the generator and other equipment.

[0196] As a preferred implementation, according to the target optimization function and the temperature transfer path model, the temperature adjustment method of the generator and other equipment is determined by a collaborative optimization algorithm, and the control strategy is generated according to the temperature adjustment method of the generator and other equipment, including the following steps:

[0197] S231, initializing the temperature adjustment method of the generator and other equipment based on the target optimization function and the temperature transfer path model;

[0198] Specifically, the initialization adjustment method must meet the physical laws of temperature transfer and the constraints of the optimization target. Get the current temperature data of the generator and other equipment:

[0199] First, the generator temperature, equipment temperature, generator target temperature and equipment target temperature are obtained, and initial temperature adjustment values ​​are generated for the generator and equipment. According to the temperature transfer path model, the initial plan is adjusted to conform to the temperature transfer law. In the collaborative optimization algorithm, the initial population is a set of possible adjustment plans.

[0200] S232, using a collaborative optimization algorithm to evaluate the performance of the current temperature adjustment method by calculating a fitness value, and generating a new temperature adjustment method through selection, crossover and mutation operations according to the evaluation result;

[0201] It should be noted that when using the collaborative optimization algorithm to optimize the temperature adjustment method of generators and other equipment, it is necessary to evaluate the performance of the current temperature adjustment method by calculating the fitness value, and generate a new temperature adjustment method through selection, crossover and mutation operations based on the evaluation results. For each individual in the population, the fitness value is calculated using the fitness function.

[0202] According to the fitness value, a part of the individuals with excellent performance are selected as parents for subsequent crossover and mutation operations. Common selection strategies include roulette selection and tournament selection. According to the selection strategy, a certain number of parent individuals are selected from the population.

[0203] Determine the crossover operation method, such as single-point crossover, double-point crossover or uniform crossover, and perform crossover operations on the selected parent individuals to generate new offspring individuals. The crossover operation aims to combine the excellent genes of the parent individuals to generate new individuals with potentially better performance;

[0204] Determine the mutation operation method, such as random mutation, non-uniform mutation, etc., perform mutation operations on offspring individuals, introduce new gene mutations, and increase the diversity of the population. Mutation operations help to jump out of the local optimal solution and explore a broader solution space;

[0205] Determine how to update the population, such as replacing some or all of the parent individuals with offspring individuals, or retaining the best individuals based on fitness values, etc., and update the individuals in the population according to the replacement strategy.

[0206] S233, repeating the evaluation, selection, crossover and mutation operations, gradually optimizing the temperature adjustment method, until the maximum number of iterations is reached, and obtaining the final temperature adjustment method;

[0207] S234. Determine the control strategy of the generator and other equipment according to the final temperature adjustment method.

[0208] Specifically, based on the temperature adjustment method, clarify the goals of the control strategy, such as maintaining the equipment within the optimal operating temperature range and reducing energy consumption. According to the temperature adjustment method, design the control logic, including when to adjust the temperature, how to adjust it, and the adjustment range. For each device, write detailed control steps, including start / stop conditions, specific instructions for temperature adjustment, etc.

[0209] S3. According to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored to obtain the sensor response data. The sensor response index is obtained by comprehensive analysis based on the sensor response data. The transmission quality evaluation value is obtained by comprehensive analysis based on the sensor response index and the environmental interference evaluation value. The control effect is evaluated based on the transmission quality evaluation value and the feedback mechanism.

[0210] As a preferred implementation, according to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored, the sensor response data is obtained, the sensor response index is obtained according to the comprehensive analysis of the sensor response data, the transmission quality evaluation value is obtained according to the comprehensive analysis of the sensor response index and the environmental interference evaluation value, and the control effect is evaluated according to the transmission quality evaluation value and the feedback mechanism, including the following steps:

[0211] S31. According to the generated control strategy, adjust the operating parameters of the generator and other equipment in real time, and continuously monitor the temperature response of the generator and other equipment;

[0212] Specifically, adjusting the operating parameters of the generator and other equipment in real time according to the generated control strategy and continuously monitoring their temperature response are key steps to ensure stable system operation and optimized performance. The following is a detailed implementation and monitoring process:

[0213] 1. Implement control strategies

[0214] Parameter setting: Input the temperature set point, adjustment rate, working schedule and other parameters determined in the control strategy into the control system of the generator and other equipment.

[0215] Automatic control: Ensure that the control system can automatically adjust the operation of the equipment according to the set parameters, such as using a PID controller (proportional-integral-differential controller) to achieve precise temperature control.

[0216] Manual / automatic switching: When necessary, provide switching function between manual and automatic control, so as to be able to respond flexibly when the automatic control system fails or requires manual intervention.

[0217] 2. Real-time monitoring of temperature response

[0218] Data Collection: Use a data acquisition system or SCADA (Supervisory Control and Data Acquisition) system to collect data from temperature sensors periodically or in real time.

[0219] Data visualization: The collected temperature data is displayed in the form of charts, curves, etc., so that operators can intuitively understand the temperature response of the equipment.

[0220] 3. Feedback mechanism to evaluate control effectiveness

[0221] Set evaluation indicators: According to the control objectives, set indicators for evaluating the control effect, such as temperature fluctuation range, energy consumption reduction percentage, equipment failure rate, etc.

[0222] Calculate feedback value: Based on the real-time monitored temperature data and other relevant parameters, calculate the actual value of the evaluation index, that is, the feedback value.

[0223] Compare and analyze: Compare the feedback value with the expected control target and analyze the implementation effect of the control strategy. If the feedback value does not reach the expected target, the control strategy needs to be further adjusted.

[0224] Adjust control strategy: Based on the evaluation results, make necessary adjustments to the control strategy, such as modifying the temperature set point, adjusting the adjustment rate, etc., to improve the control effect.

[0225] 4. Continuous monitoring and optimization

[0226] Long-term monitoring: Continuously monitor the temperature response and operating status of generators and other equipment to ensure long-term stable operation of the system.

[0227] Regular evaluation: Evaluate control effectiveness regularly (e.g. monthly, quarterly) to identify and resolve problems in a timely manner.

[0228] Optimize control strategy: Based on the results of long-term monitoring and evaluation, continuously optimize the control strategy to improve the overall performance and efficiency of the system.

[0229] 5. Emergency response mechanism

[0230] Set alarm thresholds: Set alarm thresholds for key temperature parameters. Once the temperature exceeds the threshold, the alarm is triggered immediately.

[0231] Emergency Response Procedure: Develop a detailed emergency response procedure, including alarm response, troubleshooting, emergency shutdown and other measures to ensure rapid response when equipment fails or abnormal conditions occur.

[0232] S32, the sensor response data, including response time and data transmission speed, obtains the critical response time, reference data transmission speed and allowable deviation data transmission speed from the temperature control database, and comprehensively analyzes to obtain the sensor response index;

[0233] The sensor response index is obtained as follows:

[0234]

[0235] Wherein, ξS1 represents the sensor response index, α4 represents the sensor response index influence factor corresponding to the response time, α5 represents the sensor response index influence factor corresponding to the data transmission speed, H1 represents the response time, H0 represents the critical response time, V1 represents the data transmission speed, V0 represents the reference data transmission speed, and V2 represents the allowable deviation data transmission speed.

[0236] α4 and α5 are the sensor response index influencing factors corresponding to the response time and data transmission speed preset in the temperature control database, respectively. These influencing factors are numerical indicators that measure the influence of the above sensor parameters on the sensor response index. Specifically, there is a mapping relationship table for each response time and data transmission speed, which records each possible parameter value and its corresponding sensor response index influencing factor. These mapping relationships can be one-to-one or many-to-one. For example, in practical applications, when it is necessary to evaluate the response index of a certain sensor, the measured response time and data transmission speed can be input into the corresponding mapping relationship table respectively, and the sensor response index influencing factors corresponding to these values ​​can be quickly found, where the value range of the influencing factor is between 0 and 1.

[0237] S33. Obtain a critical sensor response index and a critical environmental interference assessment value from a temperature control database, obtain a transmission quality assessment value based on a comprehensive analysis of the sensor response index, the environmental interference assessment value, the critical sensor response index and the critical environmental interference assessment value, perform evaluation feedback based on the transmission quality assessment value, and use a feedback mechanism to evaluate the control effect.

[0238] The transmission quality evaluation value is obtained in the following manner:

[0239]

[0240] Wherein, ξQ represents the transmission quality evaluation value, α6 represents the transmission quality evaluation influencing factor corresponding to the environmental interference evaluation value, α7 represents the transmission quality evaluation influencing factor corresponding to the sensor response index, ξE1 represents the environmental interference evaluation value, ξE0 represents the critical environmental interference evaluation value, ξS1 represents the sensor response index, and ξS0 represents the critical sensor response index.

[0241] α6 and α7 are the transmission quality assessment influencing factors corresponding to the environmental interference assessment value and the sensor response index preset in the temperature control database, respectively. These influencing factors are numerical indicators that measure the influence of the above parameters on the transmission quality assessment value. Specifically, there is a mapping relationship table for each environmental interference assessment value and the sensor response index, which records each possible parameter value and its corresponding transmission quality assessment influencing factor. These mapping relationships can be one-to-one or many-to-one. For example, in actual applications, when it is necessary to evaluate a certain transmission quality assessment value, the measured environmental interference assessment value and the sensor response index can be input into their respective corresponding mapping relationship tables, and the transmission quality assessment influencing factors corresponding to these values ​​can be quickly found, where the value range of the influencing factor is between 0 and 1.

[0242] As a preferred method, performing evaluation feedback according to the transmission quality evaluation value includes the following steps:

[0243] S331. Obtain the transmission quality assessment threshold from the temperature control database, compare the transmission quality assessment value with the transmission quality assessment threshold, if the transmission quality assessment value is less than the transmission quality assessment threshold, notify relevant personnel to perform maintenance, if the transmission quality assessment value is greater than or equal to the transmission quality assessment threshold, no additional processing is performed.

[0244] Specifically, the transmission quality assessment threshold is a preset value used to measure the quality of data transmission. If the transmission quality assessment value is less than the transmission quality assessment threshold, it indicates that the data transmission quality is insufficient, and problems such as inaccurate data, large delays, or even packet loss may occur. At this time, the system will notify relevant personnel to perform maintenance, such as checking sensors, communication equipment, lines or signals, to ensure that the equipment resumes normal operation and avoids affecting the system's temperature control and the safety of the wind turbine. If the transmission quality assessment value is greater than or equal to the transmission quality assessment threshold, it indicates that the transmission quality meets the standards and the system can continue to operate without additional processing.

[0245] like Figure 2 As shown, according to another embodiment of the present invention, a wind turbine generator temperature control system based on data analysis is provided, and the wind turbine generator temperature control system based on data analysis includes: a temperature transfer path analysis module 1, a control strategy generation module 2 and a control effect evaluation module 3;

[0246] Temperature transfer path analysis module 1 is used to obtain temperature data and environmental interference data of generators and other equipment in wind turbines, obtain environmental interference evaluation values ​​based on comprehensive analysis of environmental interference data, obtain temperature data correction values ​​based on matching of environmental interference evaluation values, and construct time series data based on timestamp information of temperature data and temperature data correction values; analyze the temperature transfer path between generators and other equipment based on variable mode decomposition method fused with Tiger energy operator and combined with transfer entropy theory;

[0247] The control strategy generation module 2 is used to generate a control strategy by collaboratively controlling the temperature of the generator and other devices in a generator following manner according to the temperature transfer path between the generator and other devices and in accordance with the collaborative optimization algorithm;

[0248] The control effect evaluation module 3 is used to adjust the operating parameters of the generator and other equipment in real time according to the generated control strategy, and continuously monitor the temperature response of the generator and other equipment, obtain sensor response data, obtain the sensor response index based on the comprehensive analysis of the sensor response data, obtain the transmission quality evaluation value based on the comprehensive analysis of the sensor response index and the environmental interference evaluation value, and evaluate the control effect based on the transmission quality evaluation value and the feedback mechanism.

[0249] In summary, with the help of the above technical scheme of the present invention, the present invention obtains the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtains the environmental interference evaluation value according to the comprehensive analysis of the environmental interference data, obtains the temperature data correction value according to the matching of the environmental interference evaluation value, and constructs the time series data according to the timestamp information of the temperature data and the temperature data correction value. The method can monitor the temperature changes of each device in real time and accurately, reduce the impact of environmental interference on the temperature data, and use the variable mode decomposition method integrating the Tiger energy operator and combining with the transfer entropy theory to deeply analyze the temperature transfer path between the generator and other equipment, which is helpful to understand the internal mechanism of temperature change and provide a scientific basis for subsequent temperature control. Through the collaborative optimization algorithm, the method can generate a collaborative control strategy for the generator to follow the temperature changes of other equipment. This strategy takes into account the mutual influence between the various devices, ensures the coordination and effectiveness of temperature control, and thus improves the operating efficiency of the entire wind turbine. The present invention decomposes the temperature time series data into multiple intrinsic mode function components by the variable mode decomposition method, and optimizes the processing by combining the Tiger energy operator and the singular value decomposition, effectively removes the noise and redundant information in the original data, improves the accuracy and reliability of the data, and uses the reconstructed denoised time series data, combined with the transfer entropy theory to calculate the temperature transfer entropy between the generator and other devices, and can accurately identify the temperature transfer path between the devices. By accurately calculating the temperature transfer entropy, the intensity and direction of the temperature transfer between the devices can be quantified, so as to formulate a more targeted and effective temperature control strategy. The present invention uses the generator and other devices as nodes and establishes a model according to the temperature transfer path. The method realizes the comprehensive and systematic control of the temperature of each device in the wind turbine. This systematic control method helps to ensure the temperature coordination between the devices, avoid local overheating or insufficient cooling, construct a target optimization function with the generator temperature as the core control target, and set constraints, so that the formulation of the control strategy is more targeted and clear, which helps to ensure that the temperature of the generator can be preferentially and effectively controlled during the control process, thereby ensuring the safe operation of the core components of the wind turbine.

[0250] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0251] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A wind turbine generator temperature control method based on data analysis, characterized in that: The wind turbine generator temperature control method based on data analysis includes the following steps: S1. Obtain the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtain the environmental interference evaluation value based on the comprehensive analysis of the environmental interference data, obtain the temperature data correction value based on the matching of the environmental interference evaluation value, and construct the time series data based on the timestamp information of the temperature data and the temperature data correction value; analyze the temperature transfer path between the generator and other equipment based on the variable mode decomposition method fused with the Tiger energy operator and combined with the transfer entropy theory; S2. Based on the temperature transfer path between the generator and other equipment and the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner to generate a control strategy; S3. According to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored to obtain the sensor response data. The sensor response index is obtained by comprehensive analysis based on the sensor response data. The transmission quality evaluation value is obtained by comprehensive analysis based on the sensor response index and the environmental interference evaluation value. The control effect is evaluated based on the transmission quality evaluation value and the feedback mechanism.

2. A wind turbine generator temperature control method based on data analysis according to claim 1, characterized in that: The method comprises obtaining temperature data and environmental interference data of the generator and other equipment in the wind turbine group, obtaining an environmental interference evaluation value based on a comprehensive analysis of the environmental interference data, obtaining a temperature data correction value based on matching of the environmental interference evaluation value, and constructing time series data based on the timestamp information of the temperature data and the temperature data correction value; Based on the variable mode decomposition method integrating the Tiger energy operator and combining the transfer entropy theory to analyze the temperature transfer path between the generator and other equipment, the following steps are included: S11, respectively collecting temperature data of the generator and other equipment in the wind turbine, and constructing time series data based on the timestamp information of the temperature data; S12, the environmental interference data, including wind speed, electromagnetic interference intensity and ambient temperature, obtains the critical wind speed, critical electromagnetic interference intensity, reference ambient temperature and allowable deviation ambient temperature from the temperature control database, comprehensively analyzes to obtain the environmental interference evaluation value, and obtains the temperature data correction value according to the matching of the environmental interference evaluation value; S13, using the temperature data correction value to correct the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data by the variable mode decomposition method to obtain the mode function, and optimizing each mode function by using the Tiger energy operator and singular value decomposition to obtain the reconstructed and denoised time series data; S14. Calculate the temperature transfer entropy between the generator and other devices by using the reconstructed and denoised time series data, and analyze the temperature transfer path between the generator and other devices based on the calculation result of the temperature transfer entropy.

3. A wind turbine generator temperature control method based on data analysis according to claim 2, characterized in that: The step of obtaining the temperature data correction value by matching the environmental interference evaluation value comprises the following steps: The environmental interference evaluation value is matched with the temperature data correction value corresponding to each environmental interference evaluation value preset in the temperature control database to obtain the temperature data correction value.

4. A wind turbine generator temperature control method based on data analysis according to claim 2, characterized in that: The method of using the temperature data correction value to correct the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data by the variable mode decomposition method to obtain the mode function, and optimizing each mode function by using the Tiger energy operator and singular value decomposition to obtain the reconstructed denoised time series data includes the following steps: S131, adding the temperature data correction value to the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data into multiple intrinsic mode function components by using the variable mode decomposition method, and screening the multiple intrinsic mode function components by calculating the cross-correlation and the mean square error; S132, using the Tiger energy operator to process the eigenmode function components after the screening process, by extracting the instantaneous amplitude and calculating the instantaneous power of the eigenmode function components, to obtain the instantaneous power sequence of the eigenmode function components; S133. Based on the instantaneous power sequence of the intrinsic mode function components, singular value decomposition is performed on each intrinsic mode function component, and each intrinsic mode function component processed by the singular value decomposition is reconstructed to obtain the reconstructed and denoised time series data.

5. A wind turbine generator temperature control method based on data analysis according to claim 4, characterized in that: The method of adding the temperature data correction value to the temperature data in the time series data to obtain the corrected temperature data, decomposing the corrected temperature data into a plurality of intrinsic mode function components by means of the variable mode decomposition method, and screening the plurality of intrinsic mode function components by means of cross-correlation and mean square error calculation comprises the following steps: S1311. Collect historical temperature data of generators and other equipment in the wind turbine generator set, and calculate standard temperature data of generators and other equipment in the wind turbine generator set by statistics; S1312, adding the standard temperature data and the temperature data correction value to obtain the corrected temperature data, cross-correlating each intrinsic mode function component with the corrected temperature data, and calculating the cross-correlation value and mean square error value of the intrinsic mode function component and the corrected temperature data; S1313, combining the cross-correlation value and the mean square error value into an objective function, and comparing the value of the objective function with a preset threshold, and performing screening processing on the intrinsic mode function components corresponding to the objective function that meets the preset threshold.

6. A wind turbine generator temperature control method based on data analysis according to claim 2, characterized in that: The method of calculating the temperature transfer entropy between the generator and other devices by reconstructing the denoised time series data and analyzing the temperature transfer path between the generator and other devices according to the calculation result of the temperature transfer entropy comprises the following steps: S141, configuring a delay time according to the time series data after the generator and other devices are reconstructed and de-noised, and calculating the autocorrelation function of the time series data after the generator and other devices are reconstructed and de-noised; S142, using the generator and other devices to reconstruct the autocorrelation function of the time series data after noise reduction, and calculating the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of the generator; S143, calculating the temperature transfer entropy between the generator and the other equipment using the transfer entropy theory according to the linear correlation function of the temperature data between the generator and the other equipment and the normalized autocorrelation function of the generator and in combination with the delay time; S144. Identify the temperature transfer path between the generator and other devices based on the calculated temperature transfer entropy.

7. A wind turbine generator temperature control method based on data analysis according to claim 6, characterized in that: The autocorrelation function of the time series data after noise reduction reconstructed by using the generator and other devices, the linear correlation function of the temperature data between the generator and other devices, and the normalized autocorrelation function of other devices are calculated as follows: In the formula, represents the normalized autocorrelation function of generator x; τ represents the delay time; R xx represents the autocorrelation function of generator x; ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y; R xy represents the cross-correlation function between the temperature data of the generator x and other equipment y; R yy Represents the autocorrelation function of other devices y.

8. A wind turbine generator temperature control method based on data analysis according to claim 6, characterized in that: The calculation formula for calculating the temperature transfer entropy between the generator and other devices using the transfer entropy theory based on the linear correlation function of the temperature data between the generator and other devices and the normalized autocorrelation function of the generator, combined with the delay time, is: Where, T (y→x) It represents the temperature transfer entropy value of the temperature transferred from other device y to generator x; k and l represent the embedding dimensions of the time series data of generator x and other equipment y, respectively; τ represents the delay time; ρ xy (τ) represents the linear cross-correlation function between the temperature data of generator x and other equipment y; represents the normalized autocorrelation function of the generator x.

9. The method for controlling the temperature of a wind turbine generator based on data analysis according to claim 1, characterized in that: According to the temperature transfer path between the generator and other equipment and according to the collaborative optimization algorithm, the temperature of the generator and other equipment is collaboratively controlled in a generator following manner, and the control strategy is generated, including the following steps: S21, taking the generator and other devices as nodes, and establishing a temperature transfer path model according to the temperature transfer path between the generator and other devices; S22, constructing a target optimization function with the generator temperature as the core control target, and setting corresponding constraints; S23. According to the target optimization function and the temperature transfer path model, the temperature adjustment method of the generator and other equipment is determined through a collaborative optimization algorithm, and a control strategy is generated according to the temperature adjustment method of the generator and other equipment.

10. A wind turbine generator temperature control method based on data analysis according to claim 9, characterized in that: The method of determining the temperature adjustment method of the generator and other equipment by a collaborative optimization algorithm according to the target optimization function and the temperature transfer path model, and generating a control strategy according to the temperature adjustment method of the generator and other equipment includes the following steps: S231, initializing the temperature adjustment method of the generator and other equipment based on the target optimization function and the temperature transfer path model; S232, using a collaborative optimization algorithm to evaluate the performance of the current temperature adjustment method by calculating a fitness value, and generating a new temperature adjustment method through selection, crossover and mutation operations according to the evaluation result; S233, repeating the evaluation, selection, crossover and mutation operations, gradually optimizing the temperature adjustment method, until the maximum number of iterations is reached, and obtaining the final temperature adjustment method; S234. Determine the control strategy of the generator and other equipment according to the final temperature adjustment method.

11. A wind turbine generator temperature control method based on data analysis according to claim 1, characterized in that: According to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored, the sensor response data is obtained, the sensor response index is obtained according to the comprehensive analysis of the sensor response data, the transmission quality evaluation value is obtained according to the comprehensive analysis of the sensor response index and the environmental interference evaluation value, and the control effect is evaluated according to the transmission quality evaluation value and the feedback mechanism, which includes the following steps: S31. According to the generated control strategy, the operating parameters of the generator and other equipment are adjusted in real time, and the temperature response of the generator and other equipment is continuously monitored; S32, the sensor response data, including response time and data transmission speed, obtains the critical response time, reference data transmission speed and allowable deviation data transmission speed from the temperature control database, and comprehensively analyzes to obtain the sensor response index; S33. Obtain a critical sensor response index and a critical environmental interference assessment value from a temperature control database, obtain a transmission quality assessment value based on a comprehensive analysis of the sensor response index, the environmental interference assessment value, the critical sensor response index and the critical environmental interference assessment value, perform evaluation feedback based on the transmission quality assessment value, and use a feedback mechanism to evaluate the control effect.

12. A wind turbine generator temperature control method based on data analysis according to claim 11, characterized in that: The evaluation feedback according to the transmission quality evaluation value comprises the following steps: S331. Obtain the transmission quality assessment threshold from the temperature control database, compare the transmission quality assessment value with the transmission quality assessment threshold, if the transmission quality assessment value is less than the transmission quality assessment threshold, notify relevant personnel to perform maintenance, if the transmission quality assessment value is greater than or equal to the transmission quality assessment threshold, no additional processing is performed.

13. A wind turbine generator temperature control system based on data analysis, using the wind turbine generator temperature control method based on data analysis as described in any one of claims 1 to 12, characterized in that: The wind turbine generator temperature control system based on data analysis includes: a temperature transfer path analysis module, a control strategy generation module and a control effect evaluation module; The temperature transfer path analysis module is used to obtain the temperature data and environmental interference data of the generator and other equipment in the wind turbine, obtain the environmental interference evaluation value according to the comprehensive analysis of the environmental interference data, obtain the temperature data correction value according to the matching of the environmental interference evaluation value, and construct the time series data according to the timestamp information of the temperature data and the temperature data correction value; based on the variable mode decomposition method integrating the Tiger energy operator and combining the transfer entropy theory, the temperature transfer path between the generator and other equipment is analyzed; The control strategy generation module is used to generate a control strategy by collaboratively controlling the temperature of the generator and other devices in a generator following manner according to the temperature transfer path between the generator and other devices and in accordance with the collaborative optimization algorithm; The control effect evaluation module is used to adjust the operating parameters of the generator and other equipment in real time according to the generated control strategy, and continuously monitor the temperature response of the generator and other equipment, obtain sensor response data, obtain the sensor response index based on the comprehensive analysis of the sensor response data, obtain the transmission quality evaluation value based on the comprehensive analysis of the sensor response index and the environmental interference evaluation value, and evaluate the control effect based on the transmission quality evaluation value and the feedback mechanism.

Images