Wind turbine generator fault diagnosis method and system based on SVD-SSA-LSTM

Abstract

The invention relates to the technical field of fault diagnosis, in particular to a wind turbine generator fault diagnosis method and system based on SVD-SSA-LSTM. Comprising the steps of setting a plurality of monitoring points according to equipment parameters of a wind turbine generator tower; acquiring vibration signals of each monitoring point according to a preset monitoring time node, and generating a fault feature data packet according to all the vibration signals; establishing a fault diagnosis model, and generating a fault diagnosis result according to the fault diagnosis model and the fault feature data packet; according to the method, collected vibration signals are decomposed through singular value decomposition (SVD) noise reduction, redundancy and noise components in the signals are removed, then faults of the wind turbine generator set are diagnosed through an SSA-LSTM fault diagnosis model, the accuracy of fault diagnosis of the wind turbine generator set is improved through the processing capacity of an LSTM network for time sequence data, meanwhile, the diagnosis period is shortened, and the fault diagnosis efficiency of the wind turbine generator set is improved. And the early warning efficiency of the fault risk of the wind turbine generator is improved.

Description

Technical Field

[0001] The present application relates to the technical field of fault diagnosis, and in particular to a wind turbine fault diagnosis method and system based on SVD-SSA-LSTM. Background Art

[0002] As an important part of renewable energy, wind power generation's reliability and maintenance efficiency are crucial to the stability and economy of energy supply. However, the gearbox in the wind turbine, as a key component of the transmission system, is susceptible to failure due to harsh operating environments.

[0003] Traditional fault diagnosis methods have problems such as long diagnosis cycle and low accuracy, and cannot meet the needs of modern wind power generation systems. Summary of the invention

[0004] The purpose of this application is: to solve the above technical problems, this application provides a wind turbine fault diagnosis method and system based on SVD-SSA-LSTM, aiming to improve the diagnosis accuracy of wind turbine faults and shorten the diagnosis cycle.

[0005] In some embodiments of the present application, singular value decomposition (SVD) denoising is used to decompose the collected vibration signal to remove redundant and noise components in the signal, and then the SSA-LSTM fault diagnosis model is used to diagnose the fault of the wind turbine. The LSTM network's ability to process time series data is used to improve the accuracy of wind turbine fault diagnosis, while shortening the diagnosis cycle and improving the early warning efficiency of wind turbine failure risks.

[0006] In some embodiments of the present application, multiple monitoring points are established, multiple groups of vibration signals are collected at a single monitoring time node, and all vibration signals are mutually corrected according to a preset calibration model, thereby improving the authenticity of the collected vibration signals and avoiding distortion of the collected vibration signals due to environmental disturbances, thereby affecting the fault diagnosis results.

[0007] In some embodiments of the present application, a wind turbine fault diagnosis method based on SVD-SSA-LSTM is provided, comprising:

[0008] Set multiple monitoring points according to the equipment parameters of the wind turbine tower;

[0009] Acquire the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals;

[0010] Establish a fault diagnosis model and generate fault diagnosis results based on the fault diagnosis model and fault feature data packet;

[0011] Among them, when multiple monitoring points are established, they include:

[0012] Establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point and n is the number of monitoring points.

[0013] In some embodiments of the present application, generating a fault feature data packet according to a preprocessing model includes:

[0014] Obtain the vibration signal of each monitoring point at the current monitoring time node;

[0015] Process all vibration signals according to a preset calibration model;

[0016] Generate a primary vibration signal according to the processing result;

[0017] Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result;

[0018] Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

[0019] In some embodiments of the present application, generating a primary vibration signal according to the processing result includes:

[0020] Generate vibration curves of each monitoring point based on all monitoring signals;

[0021] Generate a standard comparison curve based on the fusion results of the vibration curves;

[0022] Generate deviation evaluation values ​​between each vibration curve and the standard comparison curve in sequence;

[0023] Establish a deviation evaluation value sequence D, D = (d1, d2...di...dn), where di is the deviation evaluation value between the vibration curve of the i-th monitoring point and the standard brick comparison curve;

[0024] Generate a modified evaluation value g;

[0025]

[0026] Preset modified evaluation value threshold G1;

[0027] If g<G1, generate the first-level processing instruction;

[0028] If g>G1, generate secondary processing instructions;

[0029] A primary vibration signal is generated according to the processing instruction.

[0030] In some embodiments of the present application, when establishing a fault diagnosis model, it includes:

[0031] Establish training set data based on historical vibration data;

[0032] Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network;

[0033] Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies;

[0034] Generate the initial fitness of each sparrow individual according to the training set data;

[0035] Set the sparrow position iteration strategy according to the preset SSA optimization model;

[0036] Output the first-level fitness of each sparrow according to the iterative strategy;

[0037] Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals;

[0038] Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual;

[0039] Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

[0040] In some embodiments of the present application, the sparrow position iteration strategy is set according to a preset SSA optimization model, including:

[0041] The sparrows in the sparrow individual sequence J are divided into discoverers, joiners and vigilants;

[0042] Establish the discoverer iteration model, joiner iteration model and vigilant iteration model;

[0043] Among them, the discoverer iteration model is:

[0044]

[0045] in, is the j-th dimension parameter of the i-th sparrow in the t-th iteration, iter max is the maximum number of iterations, ɑ m represents a random number between 0 and 1, ST is the safety factor, AL is the alarm value, Q m is a random number that obeys a normal distribution, and Lm is a 1×d dimensional matrix whose elements are all 1.

[0046] In some embodiments of the present application, the joiner iteration model includes:

[0047]

[0048] Among them, Nm is the total number of sparrows, X p is the best position of the sparrow for foraging, X worst is the position of the sparrow with the worst foraging condition, and A+ satisfies A+=AT(AAT)-1, where A is a 1×d-dimensional matrix composed of random elements of 1 or -1.

[0049] In some embodiments of the present application, the alerter iteration model includes:

[0050]

[0051] in, is the center position of the entire sparrow population in the iteration, which is free from natural enemy threats, βm is a compensation control parameter that obeys the standard normal distribution, Km is a random number between -1 and 1, εm is an infinite decimal, and f i is the current sparrow’s fitness, f g is the fitness of the sparrow currently in the best foraging position, f w is the fitness of the sparrow currently in the worst foraging position.

[0052] In some embodiments of the present application, a wind turbine fault diagnosis system based on SVD-SSA-LSTM is provided, comprising:

[0053] The central control unit is used to set multiple monitoring points according to the equipment parameters of the wind turbine tower;

[0054] A monitoring unit, comprising a plurality of monitoring submodules, wherein the monitoring submodules are used to collect vibration signals at each monitoring point;

[0055] The central control unit comprises:

[0056] The first processing module is used to establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point; n is the number of monitoring points;

[0057] The second processing module is used to obtain the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals;

[0058] The third processing module is used to establish a fault diagnosis model and generate a fault diagnosis result according to the fault diagnosis model and the fault feature data packet.

[0059] In some embodiments of the present application, the second processing module is further used to:

[0060] Obtain the vibration signal of each monitoring point at the current monitoring time node;

[0061] Process all vibration signals according to a preset calibration model;

[0062] Generate a primary vibration signal according to the processing result;

[0063] Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result;

[0064] Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

[0065] In some embodiments of the present application, the third processing module is further used to:

[0066] Establish training set data based on historical vibration data;

[0067] Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network;

[0068] Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies;

[0069] Generate the initial fitness of each sparrow individual according to the training set data;

[0070] Set the sparrow position iteration strategy according to the preset SSA optimization model;

[0071] Output the first-level fitness of each sparrow according to the iterative strategy;

[0072] Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals;

[0073] Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual;

[0074] Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

[0075] Compared with the prior art, the wind turbine fault diagnosis method and system based on SVD-SSA-LSTM in the embodiment of the present application has the following beneficial effects:

[0076] The collected vibration signal is decomposed by singular value decomposition (SVD) to remove the redundant and noise components in the signal. Then the SSA-LSTM fault diagnosis model is used to diagnose the fault of the wind turbine. The LSTM network's processing ability for time series data can improve the accuracy of wind turbine fault diagnosis, shorten the diagnosis cycle, and improve the early warning efficiency of wind turbine failure risks.

[0077] By establishing multiple monitoring points, multiple groups of vibration signals are collected at a single monitoring time node, and all vibration signals are mutually corrected according to the preset calibration model, thereby improving the authenticity of the collected vibration signals and avoiding distortion of the collected vibration signals due to environmental disturbances, thereby affecting the fault diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a flow chart of a wind turbine fault diagnosis method based on SVD-SSA-LSTM in the preferred embodiment of the present application. DETAILED DESCRIPTION

[0079] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0080] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0081] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0082] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0083] like Figure 1 As shown, a wind turbine fault diagnosis method based on SVD-SSA-LSTM in a preferred embodiment of the present application includes:

[0084] Set multiple monitoring points according to the equipment parameters of the wind turbine tower;

[0085] Acquire the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals;

[0086] Establish a fault diagnosis model and generate fault diagnosis results based on the fault diagnosis model and fault feature data packet;

[0087] Among them, when multiple monitoring points are established, they include:

[0088] Establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point and n is the number of monitoring points.

[0089] Specifically, according to the equipment parameters of the wind turbine tower, multiple monitoring points are set at the horizontal and vertical positions of the tower wall, and vibration sensors are set at each monitoring point by magnetic adsorption.

[0090] Specifically, a fault feature data packet is generated according to the preprocessing model, including:

[0091] Obtain the vibration signal of each monitoring point at the current monitoring time node;

[0092] Process all vibration signals according to a preset calibration model;

[0093] Generate a primary vibration signal according to the processing result;

[0094] Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result;

[0095] Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

[0096] Specifically, the frequency domain characteristic information at the time of the fault includes, but is not limited to, parameters such as maximum value, minimum value, peak value, variance, standard deviation, root mean square, kurtosis, frequency domain mean and standard deviation.

[0097] Specifically, by constructing a trajectory matrix and performing singular value decomposition (SVD), the trend, periodic components and noise in the vibration signal are extracted to obtain the optimized time series data. The time series data is further denoised by selecting an appropriate threshold to remove the redundant and noise components in the signal, and retain and extract the fault time-frequency domain feature information.

[0098] Specifically, the SVD model uses an adaptive hard threshold selection algorithm and a soft threshold method of unequal optimal weight shrinkage when performing singular value decomposition to improve the adaptability and accuracy of signal processing.

[0099] It can be understood that in the above embodiment, singular value decomposition (SVD) noise reduction is used to decompose the collected vibration signal to remove the redundant and noise components in the signal, and then the SSA-LSTM fault diagnosis model is used to diagnose the fault of the wind turbine. The LSTM network's processing ability for time series data is used to improve the accuracy of wind turbine fault diagnosis, while shortening the diagnosis cycle and improving the early warning efficiency of wind turbine fault risks.

[0100] In a preferred embodiment of the present application, generating a primary vibration signal according to the processing result includes:

[0101] Generate vibration curves of each monitoring point based on all monitoring signals;

[0102] Generate a standard comparison curve based on the fusion results of the vibration curves;

[0103] Generate deviation evaluation values ​​between each vibration curve and the standard comparison curve in sequence;

[0104] Establish a deviation evaluation value sequence D, D = (d1, d2...di...dn), where di is the deviation evaluation value between the vibration curve of the i-th monitoring point and the standard brick comparison curve;

[0105] Generate a modified evaluation value g;

[0106]

[0107] Preset modified evaluation value threshold G1;

[0108] If g<G1, generate the first-level processing instruction;

[0109] If g>G1, generate secondary processing instructions;

[0110] A primary vibration signal is generated according to the processing instruction.

[0111] Specifically, by comparing various vibration curves, an average curve is generated as a standard comparison curve. The larger the deviation evaluation value is, the lower the overlap between the current vibration curve and the standard comparison curve is.

[0112] Specifically, the modified evaluation value threshold G1 can be set according to the historical parameters.

[0113] The first-level processing instruction means that the current vibration signal is not affected by the environment, and the vibration signal of the monitoring point with the smallest deviation evaluation value in the horizontal direction and the vibration signal of the monitoring point with the smallest deviation evaluation value in the vertical direction are selected as the first-level vibration signal.

[0114] Specifically, the second processing instruction means that the current vibration signal is disturbed by the environment during the collection process, and the vibration signals of the monitoring points with large deviations from the evaluation values ​​need to be eliminated, and a new judgment is made based on the elimination results, and the corresponding first-level vibration signal is generated based on the judgment results.

[0115] It can be understood that in the above embodiment, by establishing multiple monitoring points, multiple groups of vibration signals are collected at a single monitoring time node, and all vibration signals are mutually corrected according to a preset calibration model, thereby improving the authenticity of the collected vibration signals and avoiding distortion of the collected vibration signals due to environmental disturbances, thereby affecting the fault diagnosis results.

[0116] In a preferred embodiment of the present application, when establishing a fault diagnosis model, it includes:

[0117] Establish training set data based on historical vibration data;

[0118] Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network;

[0119] Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies;

[0120] Generate the initial fitness of each sparrow individual according to the training set data;

[0121] Set the sparrow position iteration strategy according to the preset SSA optimization model;

[0122] Output the first-level fitness of each sparrow according to the iterative strategy;

[0123] Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals;

[0124] Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual;

[0125] Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

[0126] Specifically, the parameter configuration strategy includes relevant parameters such as the number of hidden layer nodes and learning rate of the LSTM network.

[0127] Specifically, the training set data is obtained by using fault data through simulation or digital twins.

[0128] Specifically, each unit of the LSTM network consists of a forget gate, an input gate, and an output gate. Through the gating mechanism, further processing of historical information in the previous processing stage is achieved.

[0129] Specifically, the sparrow position iteration strategy is set according to the preset SSA optimization model, including:

[0130] The sparrows in the sparrow individual sequence J are divided into discoverers, joiners and vigilants;

[0131] Establish the discoverer iteration model, joiner iteration model and vigilant iteration model;

[0132] Specifically, the SSA optimization model is initialized, and relevant parameters of the SSA optimization model are set, such as population size, maximum number of iterations, fitness function, etc.

[0133] Specifically, the discoverer iteration model is:

[0134]

[0135] in, is the j-th dimension parameter of the i-th sparrow in the t-th iteration, iter max is the maximum number of iterations, ɑ m represents a random number between 0 and 1, ST is the safety factor, AL is the alarm value, Q m is a random number that obeys a normal distribution, and Lm is a 1×d dimensional matrix whose elements are all 1.

[0136] Specifically, the joiner iteration model includes:

[0137]

[0138] Among them, Nm is the total number of sparrows, X p is the best position of the sparrow for foraging, X worst is the position of the sparrow with the worst foraging condition, and A+ satisfies A+=AT(AAT)-1, where A is a 1×d-dimensional matrix composed of random elements of 1 or -1.

[0139] Specifically, the Vigilant Iteration Model includes:

[0140]

[0141] in, is the center position of the entire sparrow population in the iteration, which is free from natural enemy threats, βm is a compensation control parameter that obeys the standard normal distribution, Km is a random number between -1 and 1, εm is an infinite decimal, and f i is the current sparrow’s fitness, f gis the fitness of the sparrow currently in the best foraging position, f w is the fitness of the sparrow currently in the worst foraging position.

[0142] Specifically, for sparrow individuals with higher fitness (producers), their positions are updated according to their current positions and parameters such as warning values ​​and safety values; for sparrow individuals with lower fitness (searchers), some monitor producers and try to compete with them for food (i.e., update their own positions to be close to the producers' optimal positions), while others choose to fly to farther areas to randomly search for food to increase the possibility of finding a better solution.

[0143] Specifically, during iterative optimization, the process of fitness evaluation and position update is repeated until the maximum number of iterations is reached or other stopping conditions are met. The global optimal solution (i.e., the LSTM network parameter configuration represented by the sparrow individual with the highest fitness) is recorded and updated, and the global optimal solution is finally output as the optimized LSTM network parameter configuration.

[0144] Based on another preferred embodiment of a wind turbine fault diagnosis method based on SVD-SSA-LSTM in any of the above preferred embodiments, this preferred embodiment provides a wind turbine fault diagnosis system based on SVD-SSA-LSTM, including:

[0145] The central control unit is used to set multiple monitoring points according to the equipment parameters of the wind turbine tower;

[0146] A monitoring unit, including a plurality of monitoring submodules, wherein the monitoring submodules are used to collect vibration signals at each monitoring point;

[0147] The central control unit includes:

[0148] The first processing module is used to establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point; n is the number of monitoring points;

[0149] The second processing module is used to obtain the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals;

[0150] The third processing module is used to establish a fault diagnosis model and generate a fault diagnosis result according to the fault diagnosis model and the fault feature data packet.

[0151] Specifically, the second processing module is further used for:

[0152] Obtain the vibration signal of each monitoring point at the current monitoring time node;

[0153] Process all vibration signals according to a preset calibration model;

[0154] Generate a primary vibration signal according to the processing result;

[0155] Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result;

[0156] Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

[0157] In a preferred embodiment of the present application, the third processing module is also used for:

[0158] Establish training set data based on historical vibration data;

[0159] Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network;

[0160] Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies;

[0161] Generate the initial fitness of each sparrow individual according to the training set data;

[0162] Set the sparrow position iteration strategy according to the preset SSA optimization model;

[0163] Output the first-level fitness of each sparrow according to the iterative strategy;

[0164] Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals;

[0165] Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual;

[0166] Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

[0167] According to the first concept of the present application, the collected vibration signal is decomposed by using singular value decomposition (SVD) denoising to remove the redundancy and noise components in the signal, and then the SSA-LSTM fault diagnosis model is used to diagnose the fault of the wind turbine. The accuracy of wind turbine fault diagnosis is improved by the LSTM network's processing ability for time series data, while shortening the diagnosis cycle and improving the early warning efficiency of wind turbine fault risks.

[0168] According to the second concept of the present application, multiple monitoring points are established, multiple groups of vibration signals are collected at a single monitoring time node, and all vibration signals are mutually corrected according to a preset calibration model, thereby improving the authenticity of the collected vibration signals and avoiding distortion of the collected vibration signals due to environmental disturbances, thereby affecting the fault diagnosis results.

[0169] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A wind turbine fault diagnosis method based on SVD-SSA-LSTM, characterized in that: include: Set multiple monitoring points according to the equipment parameters of the wind turbine tower; Acquire the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals; Establish a fault diagnosis model and generate fault diagnosis results based on the fault diagnosis model and fault feature data packet; Among them, when multiple monitoring points are established, they include: Establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point and n is the number of monitoring points.

2. The wind turbine fault diagnosis method based on SVD-SSA-LSTM according to claim 1, characterized in that: Generate a fault feature data package based on the preprocessing model, including: Obtain the vibration signal of each monitoring point at the current monitoring time node; Process all vibration signals according to a preset calibration model; Generate a primary vibration signal according to the processing result; Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result; Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

3. The wind turbine fault diagnosis method based on SVD-SSA-LSTM as claimed in claim 2, characterized in that: Generate a primary vibration signal based on the processing results, including: Generate vibration curves of each monitoring point based on all monitoring signals; Generate a standard comparison curve based on the fusion results of the vibration curves; Generate deviation evaluation values ​​between each vibration curve and the standard comparison curve in sequence; Establish a deviation evaluation value sequence D, D = (d1, d2...di...dn), where di is the deviation evaluation value between the vibration curve of the i-th monitoring point and the standard brick comparison curve; Generate a modified evaluation value g; Preset modified evaluation value threshold G1; If g<G1, generate the first-level processing instruction; If g>G1, generate secondary processing instructions; A primary vibration signal is generated according to the processing instruction.

4. The wind turbine fault diagnosis method based on SVD-SSA-LSTM as claimed in claim 2, characterized in that: When establishing a fault diagnosis model, include: Establish training set data based on historical vibration data; Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network; Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies; Generate the initial fitness of each sparrow individual according to the training set data; Set the sparrow position iteration strategy according to the preset SSA optimization model; Output the first-level fitness of each sparrow according to the iterative strategy; Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals; Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual; Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

5. The wind turbine fault diagnosis method based on SVD-SSA-LSTM as claimed in claim 4, characterized in that: The sparrow position iteration strategy is set according to the preset SSA optimization model, including: The sparrows in the sparrow individual sequence J are divided into discoverers, joiners and vigilants; Establish the discoverer iteration model, joiner iteration model and vigilant iteration model; Among them, the discoverer iteration model is: in, is the j-th dimension parameter of the i-th sparrow in the t-th iteration, iter max is the maximum number of iterations, ɑ m represents a random number between 0 and 1, ST is the safety factor, AL is the alarm value, Qm is a random number that obeys the normal distribution, and Lm is a 1×d-dimensional matrix with all elements set to 1.

6. The wind turbine fault diagnosis method based on SVD-SSA-LSTM as claimed in claim 5, characterized in that: Joiner iteration model, including: Among them, Nm is the total number of sparrows, X p is the best position of the sparrow for foraging, X worst is the position of the sparrow with the worst foraging condition, and A+ satisfies A+=AT(AAT)-1, where A is a 1×d-dimensional matrix composed of random elements of 1 or -1.

7. The wind turbine fault diagnosis method based on SVD-SSA-LSTM as claimed in claim 6, characterized in that: Vigilant iteration model, including: in, is the center position of the entire sparrow population in the iteration, which is free from natural enemy threats, βm is a compensation control parameter that obeys the standard normal distribution, Km is a random number between -1 and 1, εm is an infinite decimal, and f i is the current sparrow’s fitness, f g is the fitness of the sparrow currently in the best foraging position, f w is the fitness of the sparrow currently in the worst foraging position.

8. A wind turbine fault diagnosis system based on SVD-SSA-LSTM, adopting the wind turbine fault diagnosis method based on SVD-SSA-LSTM as described in any one of claims 1 to 7, characterized in that: include: The central control unit is used to set multiple monitoring points according to the equipment parameters of the wind turbine tower; A monitoring unit, comprising a plurality of monitoring submodules, wherein the monitoring submodules are used to collect vibration signals at each monitoring point; The central control unit comprises: The first processing module is used to establish a monitoring point sequence A, A = (a1, a2...ai...an), where ai is the i-th monitoring point; n is the number of monitoring points; The second processing module is used to obtain the vibration signal of each monitoring point according to the preset monitoring time node, and generate a fault feature data packet according to all the vibration signals; The third processing module is used to establish a fault diagnosis model and generate a fault diagnosis result according to the fault diagnosis model and the fault feature data packet.

9. The wind turbine fault diagnosis system based on SVD-SSA-LSTM as claimed in claim 8, characterized in that: The second processing module is also used for: Obtain the vibration signal of each monitoring point at the current monitoring time node; Process all vibration signals according to a preset calibration model; Generate a primary vibration signal according to the processing result; Decompose the primary vibration signal according to the preset SVD model, and generate fault time-frequency domain feature information according to the decomposition result; Generate a fault feature data packet of the current monitoring time node based on the fault time-frequency domain feature information.

10. The wind turbine fault diagnosis system based on SVD-SSA-LSTM according to claim 9, characterized in that: The third processing module is also used for: Establish training set data based on historical vibration data; Establish an LSTM network and set multiple sets of parameter configuration strategies for the LSTM network; Establish a sparrow individual sequence P, P = (p1, p2...pi...pm), where a single sparrow represents a set of parameter configuration strategies, ji is the i-th sparrow individual; m is the number of parameter configuration strategies; Generate the initial fitness of each sparrow individual according to the training set data; Set the sparrow position iteration strategy according to the preset SSA optimization model; Output the first-level fitness of each sparrow according to the iterative strategy; Establish a first-level fitness sequence B, B = (b1, b2...bi...bm), where bi is the first-level fitness of the i-th sparrow individual; m is the number of sparrow individuals; Set the sparrow individual corresponding to the maximum value bmax in the first-level fitness sequence B as the target sparrow individual; Set the parameter configuration strategy corresponding to the target sparrow individual and the LSTM network to generate the fault diagnosis model.

Images