Wind turbine generator misalignment fault diagnosis method, system, equipment and medium

Abstract

The invention discloses a wind turbine generator misalignment fault diagnosis method, system and device and a medium, and belongs to the technical field of wind turbine generator state monitoring, the method comprises the steps that wind turbine generator coupler position vibration signals and rotating speed signals are collected, and the vibration signals comprise axial vibration signals and radial vibration signals; performing noise reduction processing on the vibration signal to obtain a noise-reduced vibration signal; obtaining an axis track picture, an axial order spectrum picture and a radial order spectrum picture, fusing the axis track picture, the axial order spectrum picture and the radial order spectrum picture to obtain a low-rank fusion result, carrying out dimension connection on the low-rank fusion result and the vibration signal historical time sequence characteristics after noise reduction to obtain fused new characteristics, and carrying out noise reduction on the new characteristics; performing fault diagnosis based on the fused new features; according to the method, the signal processing method and the deep learning technology are combined, the accuracy of misalignment fault diagnosis of the wind turbine generator is improved, and a certain diagnosis effect is achieved on early potential misalignment faults.

Description

Technical Field

[0001] The invention belongs to the technical field of wind turbine status monitoring, and in particular relates to a method, system, equipment and medium for diagnosing misalignment faults of wind turbines. Background Art

[0002] According to statistics, 60% of the failures in rotating machinery are related to misalignment. Misalignment failures will generate additional forces, causing equipment vibration, bearing wear, oil film instability, rotor bending and deformation, etc. The misalignment failure of a wind turbine refers to the phenomenon that the center lines of the gearbox high-speed shaft and the generator input shaft at both ends of the coupling do not coincide during rotation, resulting in unstable operation of the shaft system. The misalignment failure of a wind turbine is one of the common mechanical failures in wind power generation systems. The causes of its occurrence are complex and diverse and have a significant impact on the stability and safety of the unit's operation. In order to avoid misalignment failures, wind farms need to conduct wind turbine alignment inspections once a year. Each alignment requires 1 day of work, which seriously increases the operating costs of wind farms.

[0003] The existing online monitoring method is based on vibration signals, and performs spectrum analysis or order spectrum analysis on the axial and radial vibration signals installed on the generator input shaft, and judges the misalignment fault of the wind turbine based on the first and second harmonic characteristics and phase characteristics of the frequency. However, the wind turbine vibration data collection process is prone to noise and unit coupled vibration interference, resulting in the need to improve the diagnostic accuracy of misalignment faults, and it is difficult to judge the risk of potential misalignment faults in the early stage. Summary of the invention

[0004] The purpose of the present invention is to provide a wind turbine misalignment fault diagnosis method, system, equipment and medium, which combine signal processing methods with deep learning technology to further improve the accuracy of wind turbine misalignment fault diagnosis and have a certain diagnostic effect on early potential misalignment faults.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for diagnosing misalignment faults of a wind turbine generator set, comprising: Collecting vibration signals and speed signals of the coupling position of the wind turbine set, wherein the vibration signals include axial vibration signals and radial vibration signals; Perform noise reduction processing on the vibration signal to obtain the noise-reduced vibration signal, and extract the historical time series characteristics of the noise-reduced vibration signal; obtain the axis trajectory image, axial order spectrum image, and radial order spectrum image based on the noise-reduced vibration signal; The axial order spectrum picture is an order spectrum picture obtained by performing order analysis on the axial acceleration combined with the rotation speed signal; the radial order spectrum picture is an order spectrum picture obtained by performing order analysis on the radial acceleration combined with the rotation speed signal; Fault diagnosis is performed based on input data, wherein the input data includes an axis trajectory image, an axial order spectrum image, a radial order spectrum image, and historical time series characteristics of the vibration signal after noise reduction.

[0006] Furthermore, when the vibration signal is subjected to noise reduction processing, integrated empirical mode decomposition is used to perform noise reduction processing on the vibration signal.

[0007] Furthermore, the historical time series characteristics of the vibration signal after noise reduction include time domain characteristics and spectrum characteristics; the time domain characteristics include root mean square value, kurtosis, root amplitude and peak-to-peak value; the spectrum characteristics include sideband energy characteristics and energy ratio of the order spectrum.

[0008] Further, performing fault diagnosis based on the input data includes: The input data is fed into the trained fault diagnosis model, and the one with the highest probability is taken as the final diagnosis result.

[0009] Furthermore, the pre-trained fault diagnosis model is trained by the following steps: Build a fault diagnosis network; Softmax is used as the activation function, and the goal is to minimize the value of the multi-classification cross entropy loss function. The axis trajectory image, axial order spectrum image, radial order spectrum image and the historical time series characteristics of the denoised vibration signal are used as input. The three fault types of parallel misalignment, angular misalignment and comprehensive misalignment, each of which is divided into three levels: general, slight and severe, with a total of nine classification labels, are used as output. The parameters of the fault diagnosis network are optimized to obtain the fault diagnosis model.

[0010] Furthermore, the fault diagnosis network includes a data processing network, a multi-layer LSTM network, a Flatten layer and a multi-layer DNN network connected in sequence; the data processing network is used to fuse the axis trajectory image, the axial order spectrum image, and the radial order spectrum image to obtain a low-rank fusion result, and dimensionally regularize the low-rank fusion result and the historical time series features.

[0011] Furthermore, it also includes warning information reminders and displays based on the diagnosis results.

[0012] In a second aspect, the present invention provides a wind turbine misalignment fault diagnosis system, comprising: A signal acquisition module, used to collect vibration signals and speed signals of the coupling position of the wind turbine set, wherein the vibration signals include axial vibration signals and radial vibration signals; A vibration signal noise reduction module is used to perform noise reduction processing on the vibration signal to obtain a vibration signal after noise reduction; A comprehensive module, used for obtaining an axis trajectory image, an axial order spectrum image, a radial order spectrum image and a historical time series feature of the vibration signal after noise reduction based on the vibration signal after noise reduction; A fault diagnosis module is used to perform fault diagnosis based on the axis trajectory image, the axial order spectrum image, the radial order spectrum image and the historical time series characteristics of the vibration signal after noise reduction.

[0013] In a third aspect, the present invention provides an electronic device, comprising: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a method for diagnosing misalignment faults of a wind turbine generator set as described in any one of the first aspects of the present invention.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the method for diagnosing misalignment faults of a wind turbine set described in any one of the first aspects of the present invention is implemented.

[0015] Compared with the prior art, the present invention has at least the following beneficial technical effects: The present invention first collects the rotation speed and vibration signals, then adaptively decomposes the collected vibration signals and reconstructs the signals to reduce noise interference and coupled vibration effects. Then, the vibration axis trajectory feature image and order spectrum image are constructed and feature fusion is performed; at the same time, the long-term feature extraction of the historical features of the vibration signal is performed. Finally, the two parts of the features are effectively fused to comprehensively diagnose the misalignment fault of the wind turbine set, give the misalignment fault category and fault degree, and provide early warning information for subsequent misalignment detection and adjustment.

[0016] The provided method and system is a diagnostic method that takes into account the influence of signal noise and coupled vibration, comprehensively considers the time domain characteristics and spectrum characteristics of the vibration signal, and also considers the historical time series characteristics of the vibration signal, further increasing the accuracy of the current wind turbine misalignment fault diagnosis.

[0017] Furthermore, the EEMD method is used to adaptively decompose the collected vibration signals and reconstruct the signals to reduce the influence of noise interference and coupled vibration, which can improve the diagnostic accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 For the overall plan; Figure 2 Schematic diagram of signal noise reduction processing; Figure 3 It is a schematic diagram of comprehensive diagnosis; Figure 4 A structural block diagram of a wind turbine misalignment fault diagnosis system provided in Embodiment 2 of the present invention; Figure 5 A block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] The present invention is further described in detail below in conjunction with the accompanying drawings: Example 1 Reference Figure 1 In this embodiment, a method for diagnosing misalignment faults of a wind turbine is provided, comprising the following steps: Step S101, the signal acquisition module collects historical data, specifically including collecting the vibration signal of the wind turbine coupling position, synchronously collecting the speed signal, and collecting the fault labels corresponding to each data. The collected vibration signal at least includes the collection of two vibration signals, axial and radial, at the generator drive end to meet the subsequent comprehensive diagnostic feature extraction requirements. There are nine fault labels in total, including three fault types: parallel misalignment, angular misalignment, and comprehensive misalignment, and each fault type is divided into three levels: general, minor, and severe.

[0022] Step S102, refer to Figure 2 ,The vibration signal noise reduction module performs noise reduction processing on the vibration signal.

[0023] The noise reduction algorithm is a general technology and can adopt any one of the methods such as singular value decomposition (SVD), empirical modal decomposition (EMD), wavelet transform (WT), etc.

[0024] In some embodiments, the present invention adopts Ensemble Empirical Mode Decomposition (EEMD) as the best practice method for noise reduction processing. The specific implementation process is: perform EEMD adaptive decomposition on the axial and radial vibration signals respectively, and then calculate the multi-scale permutation entropy of the intrinsic mode function IMF components after decomposition. If the entropy value is greater than a threshold value such as 0.6, it means that there is a lot of noise in the vibration signal and it needs to be deleted; retain the IMF components less than the threshold value. Add the filtered IMF components to obtain the vibration signal after noise reduction.

[0025] Step S103, refer to Figure 3 ,The comprehensive module obtains the axis trajectory image, the axial order spectrum image, and the radial order spectrum image based on the ,noise-reduced vibration signal, and extracts the historical time series features of the ,noise-reduced vibration signal.

[0026] First, read the vibration signal after noise reduction in step S102 and perform order analysis and key curve plotting. The plotted curve graphs include axis trajectory images, axial order spectrum images, and radial order spectrum images, where the axis trajectory image is a picture drawn with axial acceleration as the horizontal coordinate and radial acceleration as the vertical coordinate; the drawing of the order spectrum requires order analysis of the axial and radial vibration signals to obtain order spectra, and the order spectrum is plotted to obtain axial order spectrum graphs and radial order spectrum graphs.

[0027] Step S104, model training module, next, pre-process the image data, read the axial trajectory image, axial order spectrum image, and radial order spectrum image obtained in the previous step respectively, convert the color axial trajectory image, axial order spectrum image, and radial order spectrum image into grayscale images, and perform dimensionality increase processing to facilitate low-rank fusion.

[0028] Axis trajectory picture: Grayscale conversion to Dimension expansion is performed as

[0029] Axial order spectrum image: Grayscale conversion to Dimension expansion is performed as

[0030] Radial order spectrum image: Grayscale conversion to Dimension expansion is performed as

[0031] in, , , They are the original color images of the axis trajectory image, axial order spectrum image, and radial order spectrum image respectively; L and H are the sizes of the axis trajectory image, L and K are the sizes of the axial order spectrum image, and L and S are the sizes of the radial order spectrum image.

[0032] Then, the three types of image data after dimensionality increase are fused to obtain the low-rank fusion result; the dimension after fusion is set to R, and the intermediate representation dimension is set to T, then the low-rank fusion result It is expressed as:

[0033] in, , , is the weight matrix, represents matrix multiplication, Indicates phase multiplication.

[0034] Will Permuting the first and second dimensions of .

[0035] Will Perform linear transformation to obtain low-rank fusion results .

[0036]

[0037] in, is the weight matrix, and bias is the bias term. The above weight matrices and bias terms are determined after model training.

[0038] Then, the historical time series features are extracted from the axial and radial vibration signals after noise reduction respectively (the time series features can be understood as the time domain features and frequency domain features of the vibration signals extracted from the historical data of the past 10 to 15 days). The extracted historical time series features include time domain features and spectrum features. The time domain features include root mean square value, kurtosis, root amplitude, peak-to-peak value, etc.; the spectrum features include the sideband energy characteristics and energy ratio of the order spectrum. Selecting the historical time series features for a period of time as a reference for the current state can effectively memorize the historical fault degree and thus improve the accuracy of diagnosis. The historical time series features are expressed as , where S is the length of the historical time series data, such as 10 days, indicating that the current state is related to the state of the past 10 days, providing historical fault degree information to correct the current diagnosis effect; M is the number of features extracted from the historical time series features, that is, the sum of the time domain features and the frequency domain features. In order to unify the dimensions with those after low-rank fusion, it is necessary to input the historical time series features into the LSTM network for dimensional regularization. The output dimension of the LSTM network is set to R, and the historical time series features are expressed as After passing through the LSTM network, it becomes The matrix of .

[0039] Finally, the low-rank fusion features are combined with the historical time series features Perform dimension connection to form a new fused feature Z: The fused new features are automatically extracted through a multi-layer LSTM network to obtain feature dependencies of different time series. The Flatten layer is then used for dimension regularization and then input into a multi-layer DNN network. The last layer of the DNN network is the classification result layer, and softmax is used as the activation function. After the entire network structure is built, the axis trajectory image, axial order spectrum image, radial order spectrum image and historical time series features of the vibration signal after noise reduction obtained based on the historical speed signal and vibration signal are used as input. The three fault types of parallel misalignment, angular misalignment and comprehensive misalignment are used, and each type is divided into three levels of general, slight and severe, with a total of nine classification labels as output. The designed network structure is used for feature extraction, and the multi-classification cross entropy loss function is used as the target to optimize the parameters of the designed network structure to measure the optimal fault diagnosis model.

[0040] Step S105, fault diagnosis module, the data to be predicted is processed in the same data processing method as step S101 to step S103 to meet the data input format of the optimal fault diagnosis model, and the trained optimal fault diagnosis model is used to perform diagnostic analysis on the data to be predicted to obtain the diagnosis result. Among them, the data to be predicted includes the position vibration signal and speed signal of the coupling of the wind turbine, and the vibration signal includes at least the collection of the axial and radial vibration signals of the generator drive end.

[0041] The diagnostic results are divided into three categories: parallel misalignment, angular misalignment, and comprehensive misalignment. Each category is divided into three levels: general, slight, and severe. Therefore, it is a 9-category diagnostic task. After the diagnostic result module completes the diagnosis, it will issue an early warning message and display the diagnostic results to prompt maintenance personnel to verify and maintain.

[0042] The following is a system embodiment of the present invention, which can be used to implement the method embodiment of the present invention. For details not disclosed in the system embodiment, please refer to the method embodiment of the present invention.

[0043] Example 2 See also Figure 4 In yet another embodiment of the present invention, a wind turbine misalignment fault diagnosis system is provided, comprising: A signal acquisition module, used to collect vibration signals and speed signals of the coupling position of the wind turbine set, wherein the vibration signals include axial vibration signals and radial vibration signals; A vibration signal noise reduction module is used to perform noise reduction processing on the vibration signal to obtain a vibration signal after noise reduction; A comprehensive module, used for obtaining an axis trajectory image, an axial order spectrum image, a radial order spectrum image and a historical time series feature of the vibration signal after noise reduction based on the vibration signal after noise reduction; A fault diagnosis module is used to perform fault diagnosis based on the axis trajectory image, the axial order spectrum image, the radial order spectrum image and the historical time series characteristics of the vibration signal after noise reduction.

[0044] All relevant contents of each step involved in the aforementioned embodiment of a method for diagnosing misalignment faults of wind turbines can be referred to the functional description of the functional modules corresponding to a system for diagnosing misalignment faults of wind turbines in the embodiment of the present invention, and will not be repeated here.

[0045] Example 3 Reference Figure 5, an embodiment of the present invention provides an electronic device, which includes a processor and a memory, and the processor and the memory are connected through a bus; the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a method for diagnosing misalignment faults of a wind turbine. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.

[0046] Example 4 The embodiment of the present invention provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the computer-readable storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of a method for diagnosing misalignment faults of a wind turbine in the above embodiment.

[0047] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0048] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0049] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0050] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0051] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for diagnosing misalignment faults of a wind turbine generator set, characterized in that: include: Collecting vibration signals and speed signals of the coupling position of the wind turbine set, wherein the vibration signals include axial vibration signals and radial vibration signals; Perform noise reduction processing on the vibration signal to obtain the noise-reduced vibration signal, and extract the historical time series characteristics of the noise-reduced vibration signal; obtain the axis trajectory image, axial order spectrum image, and radial order spectrum image based on the noise-reduced vibration signal; The axial order spectrum picture is an order spectrum picture obtained by performing order analysis on the axial acceleration combined with the rotation speed signal; the radial order spectrum picture is an order spectrum picture obtained by performing order analysis on the radial acceleration combined with the rotation speed signal; Fault diagnosis is performed based on input data, wherein the input data includes an axis trajectory image, an axial order spectrum image, a radial order spectrum image, and historical time series characteristics of the vibration signal after noise reduction.

2. A wind turbine misalignment fault diagnosis method according to claim 1, characterized in that: When the vibration signal is subjected to noise reduction processing, integrated empirical mode decomposition is used to perform noise reduction processing on the vibration signal.

3. A wind turbine misalignment fault diagnosis method according to claim 1, characterized in that: The historical time series characteristics of the vibration signal after noise reduction include time domain characteristics and spectrum characteristics; the time domain characteristics include root mean square value, kurtosis, root mean square amplitude and peak-to-peak value; The spectrum characteristics include the sideband energy characteristics and energy ratio of the order spectrum.

4. A wind turbine misalignment fault diagnosis method according to claim 1, characterized in that: Performing fault diagnosis based on the input data includes: The input data is fed into the pre-trained fault diagnosis model, and the one with the highest probability is taken as the final diagnosis result.

5. A wind turbine misalignment fault diagnosis method according to claim 4, characterized in that: The pre-trained fault diagnosis model is obtained by training in the following steps: Build a fault diagnosis network; Softmax is used as the activation function, and the goal is to minimize the value of the multi-classification cross entropy loss function. The axis trajectory image, axial order spectrum image, radial order spectrum image and the historical time series characteristics of the denoised vibration signal are used as input. The three fault types of parallel misalignment, angular misalignment and comprehensive misalignment, each of which is divided into three levels: general, slight and severe, with a total of nine classification labels, are used as output. The parameters of the fault diagnosis network are optimized to obtain the fault diagnosis model.

6. A wind turbine misalignment fault diagnosis method according to claim 5, characterized in that: The fault diagnosis network includes a data processing network, a multi-layer LSTM network, a Flatten layer and a multi-layer DNN network connected in sequence; the data processing network is used to fuse the axis trajectory image, the axial order spectrum image, and the radial order spectrum image to obtain a low-rank fusion result, and dimensionally regularize the low-rank fusion result and the historical time series features.

7. A wind turbine misalignment fault diagnosis method according to claim 1, characterized in that: It also includes early warning information reminders and displays based on the diagnosis results.

8. A wind turbine misalignment fault diagnosis system, characterized in that: include: A signal acquisition module, used to collect vibration signals and speed signals of the coupling position of the wind turbine set, wherein the vibration signals include axial vibration signals and radial vibration signals; A vibration signal noise reduction module is used to perform noise reduction processing on the vibration signal to obtain a vibration signal after noise reduction; A comprehensive module, used for obtaining an axis trajectory image, an axial order spectrum image, a radial order spectrum image and a historical time series feature of the vibration signal after noise reduction based on the vibration signal after noise reduction; A fault diagnosis module is used to perform fault diagnosis based on the axis trajectory image, the axial order spectrum image, the radial order spectrum image and the historical time series characteristics of the vibration signal after noise reduction.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a wind turbine misalignment fault diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a method for diagnosing misalignment faults of a wind turbine set according to any one of claims 1 to 7 is implemented.

Images