Fault diagnosis method, device and equipment for transmission chain of wind turbine generator and medium
By demarcating molecular cycles in the transmission chain of the wind turbine unit and resampling the vibration signal, combined with spectrum analysis and evaluation of pre-trained models, the problem of low fault diagnosis accuracy in the prior art is solved, and more efficient and accurate fault diagnosis is achieved.
Patent Information
- Application Number
- CN202510007262.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-03
- Publication Date
- 2025-04-15
AI Technical Summary
The fault diagnosis method of existing wind turbine drive chains relies on a single root mean square index, resulting in low diagnostic accuracy.
The rotational speed pulse signal during the sampling period divides the molecular cycle, resamples the vibration signal, determines the spectrum, and inputs the target spectrum into the pre-trained transmission chain state evaluation model to evaluate the status of the transmission chain and performs troubleshooting.
It improves the accuracy and efficiency of fault diagnosis of the transmission chain of the wind turbine unit, simplifies the data processing process, reduces calculation costs, and makes the fault diagnosis results more accurate.
Smart Images
Figure CN119825652A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present invention relate to the technical field of wind power generation, and in particular to a fault diagnosis method, device, equipment and medium for a transmission chain of a wind turbine generator set. Background Art
[0002] As the global demand for clean energy continues to grow, wind power generation, as one of the important renewable energy sources, has been widely used. In the wind power generation system, the wind turbine drive chain plays a key role in converting wind energy into electrical energy, and its stable operation is essential to ensure the efficiency and reliability of wind power generation. Therefore, accurately diagnosing the fault of the wind turbine drive chain is an important link to ensure the stable operation of the wind turbine drive chain.
[0003] At present, the fault diagnosis method of the wind turbine transmission chain is mainly realized by calculating the root mean square (RMS) index of the vibration signal. However, relying on a single type of monitoring index to determine the fault has the problem of low accuracy.
[0004] Therefore, it is urgent to propose a new method to solve the above problems. Summary of the invention
[0005] The present invention provides a fault diagnosis method, device, equipment and medium for a transmission chain of a wind turbine generator set, which can improve the accuracy of the fault diagnosis of the transmission chain of the wind turbine generator set.
[0006] In a first aspect, an embodiment of the present invention provides a method for diagnosing a fault in a transmission chain of a wind turbine generator set, comprising:
[0007] After dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, the resampling time of the vibration signal of the generator shaft within each of the sub-periods is determined according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain;
[0008] After determining the vibration signal corresponding to each of the resampling moments, determining the frequency spectrum of the vibration signal according to the vibration signal corresponding to each of the resampling moments;
[0009] Determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain;
[0010] A fault diagnosis result of the transmission chain is determined according to the state evaluation score.
[0011] The technical solution of the present invention is to divide the sampling period into multiple sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, and then determine the resampling time of the vibration signal of the generator shaft within each sub-period according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain; after determining the vibration signal corresponding to each resampling time, determine the spectrum of the vibration signal according to the vibration signal corresponding to each resampling time; determine the target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain the state evaluation score of the transmission chain; determine the fault diagnosis result of the transmission chain according to the state evaluation score. The above technical solution is to divide the sampling period into multiple sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, and then determine the resampling time of the vibration signal of the generator shaft within each sub-period according to the speed of the speed pulse signal, so that the number of vibration signal samples within each sub-period can be relatively uniform, thereby simplifying the subsequent data processing process, improving the efficiency of vibration signal processing, and reducing the calculation cost. And make the vibration signal within each sub-period more regular, which is helpful to improve the accuracy of subsequent fault diagnosis. After determining the vibration signal corresponding to each resampling moment, the spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling moment, which not only reduces the amount of data to be processed subsequently, but also improves the accuracy of the determined spectrum, thereby improving the accuracy of subsequent fault diagnosis. Then, the target spectrum is determined according to the sampling time of the spectrum and the speed pulse signal, and the vibration characteristics and the speed characteristics can be integrated, avoiding the one-sidedness caused by relying on only a single piece of information to evaluate the state of the transmission chain, so that the accuracy of the subsequent transmission chain fault diagnosis results is higher. After that, the target spectrum is input into the pre-trained transmission chain state evaluation model to obtain the state evaluation score of the transmission chain, which can achieve rapid acquisition of the state evaluation score and improve the efficiency of subsequent diagnosis. Finally, the fault diagnosis result of the transmission chain is determined according to the state evaluation score, which can accurately determine whether the transmission chain has a fault.
[0012] In a second aspect, an embodiment of the present invention further provides a fault diagnosis device for a transmission chain of a wind turbine generator set, the device comprising:
[0013] A resampling module, for dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, and determining the resampling time of the vibration signal of the generator shaft within each sub-period according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain;
[0014] A vibration spectrum determination module, configured to determine the spectrum of the vibration signal according to the vibration signal corresponding to each resampling moment after determining the vibration signal corresponding to each resampling moment;
[0015] An evaluation score determination module is used to determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain;
[0016] A diagnosis module is used to determine a fault diagnosis result of the transmission chain according to the state evaluation score.
[0017] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:
[0018] at least one processor; and a memory communicatively coupled to the at least one processor;
[0019] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can implement the fault diagnosis method for the transmission chain of a wind turbine set as described in any one of the first aspects.
[0020] In a fourth aspect, an embodiment of the present invention further provides a storage medium including computer executable instructions,
[0021] The computer executable instructions, when executed by a computer processor, implement any method for diagnosing a fault in a transmission chain of a wind turbine generator set as described in the first aspect.
[0022] It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer-readable storage medium may be packaged together with the processor of the fault diagnosis device of the transmission chain of the wind turbine generator set, or may be packaged separately from the processor of the fault diagnosis device of the transmission chain of the wind turbine generator set, and this application does not limit this.
[0023] The description of the second, third and fourth aspects in this application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third and fourth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.
[0024] In this application, the name of the fault diagnosis device for the transmission chain of the wind turbine set does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they belong to the scope of the claims of this application and their equivalent technologies.
[0025] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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.
[0027] Figure 1 A flow chart of a fault diagnosis method for a transmission chain of a wind turbine generator set provided by an embodiment of the present invention;
[0028] Figure 2 A flow chart of another method for diagnosing a fault in a transmission chain of a wind turbine generator set provided by an embodiment of the present invention;
[0029] Figure 3 A schematic diagram of the structure of a fault diagnosis device for a transmission chain of a wind turbine generator set provided by an embodiment of the present invention;
[0030] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.
[0032] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.
[0033] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.
[0034] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application 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 limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.
[0035] It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe various operations (or steps) as sequential processes, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.
[0036] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.
[0037] In the description of the present application, unless otherwise specified, “plurality” means two or more.
[0038] Figure 1 A flow chart of a fault diagnosis method for a transmission chain of a wind turbine set provided in an embodiment of the present invention. This embodiment is applicable to a situation where a fault diagnosis of a transmission chain of a wind turbine set is required. The method can be performed by a fault diagnosis device for a transmission chain of a wind turbine set, and specifically includes the following steps:
[0039] Step 110, after dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator within the sampling period, determine the resampling time of the vibration signal of the generator shaft within each sub-period according to the speed of the speed pulse signal.
[0040] Among them, a wind turbine refers to a device that converts wind energy into electrical energy. A wind turbine includes a generator shaft and a transmission chain. The sampling period refers to a time interval for sampling relevant signals of the wind turbine (such as a speed pulse signal) during the signal acquisition process. The speed pulse signal refers to a signal related to the speed of the generator shaft. The sub-period refers to a smaller time interval divided according to the sampling time of the speed pulse signal within the sampling period. The vibration signal of the generator shaft refers to a signal formed by the vibration generated by the mechanical movement or external excitation of the generator shaft during operation. The resampling time refers to the time point for sampling the vibration signal that is re-determined according to the speed of the speed pulse signal in each sub-period.
[0041] Specifically, the sampling period is first divided into multiple sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine set within the sampling period. For example, if the sampling period is 16:35:00-16:35:10, and the sampling times of the speed pulse signal within the sampling period are 16:35:02, 16:35:04, 16:35:06, 16:35:08 and 16:35:10, then the sampling period can be divided into 5 sub-periods (16:35:00-16:35:02, 16:35:02-16:35:04, 16:35:04-16:35:06, 16:35:06-16:35:08 and 16:35:08-16:35:10), and the duration of each sub-period is 2 seconds.
[0042] After determining a plurality of sub-periods, the lowest speed corresponding to the speed pulse signal may be determined according to the speed pulse signal, and then the resampling time of the vibration signal in each sub-period may be determined according to the lowest speed.
[0043] In this embodiment, the sampling period is divided into multiple sub-periods according to the sampling time of the speed pulse signal, and the resampling time in each sub-period is determined according to the speed of the speed pulse signal, so that equal-angle sampling can be achieved, that is, in each sub-period, the vibration signal samples are collected when the generator shaft rotates at the same angle. Equal-angle sampling makes the number of vibration signal samples in each sub-period relatively uniform, thereby simplifying the subsequent data processing process, improving the efficiency of vibration signal processing, and reducing the calculation cost. In addition, the vibration signal after equal-angle sampling is more regular, which helps to improve the accuracy of subsequent fault diagnosis.
[0044] Step 120: After the vibration signal corresponding to each resampling moment is determined, the frequency spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling moment.
[0045] The frequency spectrum of the vibration signal refers to a frequency distribution diagram obtained by performing a spectrum analysis on the vibration signal.
[0046] Specifically, after determining the resampling time of the vibration signal of the generator shaft in each sub-period, the vibration signal at each resampling time can be determined based on the vibration signal at the original sampling time in the sampling period. Specifically, for the original sampling time that directly corresponds to the resampling time, the vibration signal at the original sampling time can be directly used as the vibration signal at the resampling time. For the resampling time that does not directly correspond to any original sampling time, the vibration signal at the original sampling time can be processed using an interpolation method (such as linear interpolation or polynomial interpolation) to obtain the vibration signal at the remaining resampling time.
[0047] Then, according to the actual situation or needs, a suitable Fourier transform algorithm (such as fast Fourier transform or short-time Fourier transform) is selected, and then the selected Fourier transform algorithm is applied to process the vibration signal corresponding to each resampling moment to obtain the spectrum of the vibration signal.
[0048] In this embodiment, the frequency spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling moment, which not only reduces the amount of data to be subsequently processed, but also improves the accuracy of the determined frequency spectrum, thereby improving the accuracy of subsequent fault diagnosis.
[0049] Step 130 : determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state assessment model to obtain a transmission chain state assessment score.
[0050] The target spectrum refers to the spectrum used to evaluate the state of the transmission chain, which is determined based on the sampling time of the spectrum and the speed pulse signal. The transmission chain state evaluation model refers to a pre-trained deep learning model, which can evaluate the state of the transmission chain based on the input target spectrum information and output a state evaluation score. The state evaluation score refers to a numerical value output by the transmission chain state evaluation model, which is used to quantitatively represent the state of the transmission chain.
[0051] Specifically, the spectrum can be normalized according to the ratio of the speed corresponding to each frequency in the spectrum to the rated speed of the generator shaft to obtain the normalized spectrum of the vibration signal. Then, the normalized spectrum and the sampling time of the speed pulse signal are fused based on weighted average, maximum value selection or minimum value selection to obtain the target spectrum. Finally, the target spectrum is input into the pre-trained transmission chain state assessment model to obtain the transmission chain state assessment score.
[0052] In this embodiment, the target spectrum is determined according to the sampling time of the spectrum and the speed pulse signal, and the vibration characteristics and the speed characteristics can be integrated, avoiding the one-sidedness caused by relying on only a single piece of information to evaluate the state of the transmission chain, so that the accuracy of the subsequent transmission chain fault diagnosis results is higher. After that, the target spectrum is input into the pre-trained transmission chain state evaluation model to obtain the state evaluation score of the transmission chain, which can achieve rapid state evaluation score, and improve the efficiency of subsequent diagnosis.
[0053] Step 140: Determine a fault diagnosis result of the transmission chain according to the status evaluation score.
[0054] The fault diagnosis result refers to the conclusion on whether there is a fault in the transmission chain determined based on the status evaluation score.
[0055] Specifically, a preset fault threshold can be set according to the actual situation or demand, and then when the state evaluation score is greater than the preset fault threshold, the fault diagnosis result of the transmission chain is determined to be a fault, and when the state evaluation score is not greater than the preset fault threshold, the fault diagnosis result of the transmission chain is determined to be no fault. For example, the preset fault threshold can be the minimum loss function value recorded in the process of training the transmission chain state evaluation model.
[0056] In this embodiment, the fault diagnosis result is determined by the state evaluation score, so as to accurately determine whether the transmission chain has a fault.
[0057] In the embodiment of the present invention, after the sampling period is divided into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, the resampling time of the vibration signal of the generator shaft within each sub-period is determined according to the speed of the speed pulse signal, so that the number of vibration signal samples within each sub-period can be relatively uniform, thereby simplifying the subsequent data processing process, improving the efficiency of vibration signal processing, and reducing the calculation cost. And making the vibration signal within each sub-period more regular, it is helpful to improve the accuracy of subsequent fault diagnosis. After determining the vibration signal corresponding to each resampling time, the spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling time, which not only reduces the amount of data to be processed subsequently, but also improves the accuracy of the determined spectrum, thereby improving the accuracy of subsequent fault diagnosis. Then, the target spectrum is determined according to the sampling time of the spectrum and the speed pulse signal, and the vibration characteristics and the speed characteristics can be integrated, avoiding the one-sidedness caused by relying only on a single information to evaluate the state of the transmission chain, so that the accuracy of the subsequent transmission chain fault diagnosis result is higher. Then, the target spectrum is input into the pre-trained transmission chain state evaluation model to obtain the state evaluation score of the transmission chain, which can achieve rapid acquisition of the state evaluation score and improve the efficiency of subsequent diagnosis. Finally, the fault diagnosis result of the transmission chain is determined according to the status evaluation score, which can accurately determine whether the transmission chain has a fault.
[0058] Figure 2 A flowchart of another method for diagnosing a fault in a transmission chain of a wind turbine generator system provided by an embodiment of the present invention is a specific embodiment based on the above embodiment. In this embodiment, the method may further include:
[0059] Step 210: Divide the sampling period into a plurality of sub-periods according to the sampling moments of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period.
[0060] Step 211: Determine the resampling time of the vibration signal in each sub-period according to the lowest speed corresponding to the speed pulse signal in the sampling period.
[0061] Specifically, the lowest speed corresponding to the speed pulse signal within the sampling period can be first determined based on the speed pulse signal within the sampling period, and then the correspondence table between the lowest speed and the optimal sampling frequency can be queried based on the lowest speed to obtain the optimal sampling frequency corresponding to the lowest speed. Then, the resampling proportional factor can be calculated based on the ratio of the optimal sampling frequency to the current sampling frequency corresponding to the lowest speed. Then, the number of samples of each sub-period can be determined based on the duration of each sub-period and the resampling proportional factor. Finally, the duration of each sub-period can be evenly divided based on the number of samples of each sub-period to obtain the resampling moment of each sub-period.
[0062] It should be noted that the correspondence table between the minimum rotation speed and the optimal sampling frequency is set in advance according to actual conditions or needs.
[0063] In this embodiment, the resampling time of the vibration signal in each sub-period is determined according to the lowest speed corresponding to the speed pulse signal in the sampling period, thereby avoiding the situation where the collected data is either too sparse to reflect key information or too dense to cause data redundancy due to the fixed sampling frequency, and ensuring that the signal can be processed at the optimal sampling rate in each sub-period, thereby improving the validity and accuracy of the data.
[0064] Furthermore, step 211 may specifically include: determining a resampling scale factor based on the minimum speed; for the current sub-cycle, determining the number of samples of the current sub-cycle based on the duration of the current sub-cycle and the resampling scale factor; and evenly dividing the duration of the current sub-cycle based on the number of samples to obtain the resampling moment of the current sub-cycle.
[0065] The resampling scale factor refers to a coefficient determined according to the minimum speed for calculating the resampling time. The sampling quantity refers to the number of vibration signal samples to be collected in the current sub-period, calculated according to the sub-period duration and the resampling scale factor.
[0066] Specifically, the product of the minimum speed and a preset multiple (such as 2) can be calculated to obtain the resampling scale factor. The preset multiple is used to ensure that the resampled frequency is at least twice the highest frequency of the vibration signal. Then, for the current sub-period, the ratio of the duration of the current sub-period and the resampling scale factor can be calculated to obtain the number of samples of the current sub-period, and then the duration of the current sub-period is evenly divided based on the number of samples to obtain the resampling time of the current sub-period.
[0067] In this embodiment, through the above steps, the original vibration signal can be converted into an angular domain signal that is more suitable for subsequent analysis, which helps to simplify the signal processing process and reduce unnecessary calculation steps and data processing time.
[0068] Step 212: After the vibration signal corresponding to each resampling moment is determined, the frequency spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling moment.
[0069] Step 213 , normalizing the spectrum according to the ratio of the rotation speed corresponding to each frequency in the spectrum to the rated rotation speed of the generator shaft to obtain a normalized spectrum of the vibration signal.
[0070] The normalized spectrum refers to the spectrum obtained after normalizing the spectrum of the vibration signal.
[0071] Exemplarily, if the sampling points of the frequency spectrum of the vibration signal include point A (a1, a2), point B (b1, b2), point C (c1, c2) and point D (d1, d2), the rated speed of the generator shaft is v, the speed corresponding to frequency a1 is v1, the speed corresponding to frequency b1 is v2, the speed corresponding to frequency c1 is v3, and the speed corresponding to frequency d1 is v4, then the spectrum is normalized, and the sampling points of the normalized spectrum of the vibration signal include point A (a1 / (v / v1), a2), point B (b1 / (v / v2), b2), point C (c1 / (v / v3), c2) and point D (d1 / (v / v4), d2).
[0072] In this embodiment, the spectrum is normalized according to the ratio of the rotation speed corresponding to each frequency in the spectrum to the rated rotation speed of the generator shaft, thereby improving the accuracy of the target spectrum determined subsequently.
[0073] Step 215: Determine the target spectrum according to the normalized spectrum and the sampling time of the speed pulse signal.
[0074] Specifically, the normalized spectrum and the sampling time of the rotation speed pulse signal can be fused based on methods such as weighted averaging, maximum value selection or minimum value selection to obtain the target spectrum.
[0075] In this embodiment, the target spectrum is determined according to the sampling time of the normalized spectrum and the speed pulse signal, so that the vibration characteristics and the speed characteristics can be better integrated, so that the accuracy of the transmission chain fault diagnosis result determined subsequently is higher.
[0076] Furthermore, step 215 may specifically include: determining the frequency points of the normalized spectrum as the frequency points of the target spectrum; determining the frequency difference multiples according to the adjacent frequency differences of the normalized spectrum and the indexes corresponding to the frequency points of the target spectrum; determining the amplitudes corresponding to the frequency points of the target spectrum according to the frequency difference multiples, the intervals between the sampling moments of the speed pulse signal and the amplitudes corresponding to the frequency points of the normalized spectrum; constructing the target spectrum according to the frequency points of the target spectrum and the amplitudes corresponding to the frequency points of the target spectrum.
[0077] The adjacent frequency difference refers to the frequency difference between two adjacent frequency points in the spectrum. In this embodiment, the index refers to the number identifying each frequency point. For example, the index of the second frequency point of the normalized spectrum is 2.
[0078] Specifically, the frequency points of the normalized spectrum are first determined as the frequency points of the target spectrum, and the frequency difference between two adjacent frequency points in the normalized spectrum is calculated to obtain the adjacent frequency difference of the normalized spectrum, that is, the first adjacent frequency difference. At the same time, the interval between each sampling moment of the speed pulse signal is determined as the first sampling interval.
[0079] Then, for the current frequency point of the target spectrum, the product of the index corresponding to the current frequency point and the first adjacent frequency difference is calculated to obtain the first frequency difference multiple, the product of the index of the next frequency point corresponding to the current frequency point and the first adjacent frequency difference is calculated to obtain the second frequency difference multiple, and the product of the index corresponding to the current frequency point and the first sampling interval is calculated to obtain the frequency-time product. When the frequency-time product is not less than the first frequency difference multiple, and the frequency-time product is less than the second frequency difference multiple, the amplitude (f) of the current frequency point is determined according to the amplitude (f1) of the frequency point of the normalized spectrum corresponding to the current frequency point, the amplitude (f2) of the next frequency point of the normalized spectrum corresponding to the current frequency point, the first adjacent frequency difference (r1), the first frequency difference multiple (n1) and the frequency-time product (n2), and the specific calculation formula is as follows: f = f1 + (f2-f1) / r1×(n2-n1). When the frequency-time product is less than the first frequency difference multiple or the frequency-time product is not less than the second frequency difference multiple, the amplitude (f3) of the frequency point of the normalized spectrum corresponding to the current frequency point is determined as the amplitude of the current frequency point, that is, f=f3.
[0080] In this embodiment, through the above steps, the amplitude of the target spectrum is made closer to the actual operating condition, the accuracy of the determined target spectrum is improved, and thus the accuracy and reliability of subsequent fault diagnosis are significantly improved.
[0081] Step 215: input the target frequency spectrum into a pre-trained transmission chain state assessment model to obtain a transmission chain state assessment score.
[0082] Furthermore, the training process of the transmission chain state assessment model includes: obtaining the historical target spectrum of the transmission chain of each wind turbine and the root mean square index corresponding to the historical target spectrum; using the historical target spectrum of the transmission chain of each wind turbine and the root mean square index corresponding to the historical target spectrum as training data to train the generative adversarial network model to obtain the transmission chain state assessment model.
[0083] The historical target spectrum refers to the target spectrum data corresponding to the transmission chain of the wind turbine in the past period of time. The root mean square index refers to an index used to describe the fluctuation or energy level of a set of data. The generative adversarial network model refers to a deep learning model consisting of two main parts: the generator and the discriminator.
[0084] Specifically, the historical target spectrum of the transmission chain of each wind turbine is first obtained, and then the root mean square index corresponding to each historical target spectrum can be calculated using the root mean square index calculation formula. The specific calculation formula is as follows: Among them, RMS represents the root mean square index, M represents the total amplitude of the target spectrum, and f m Indicates the amplitude of the mth frequency point of the target spectrum.
[0085] These data are then cleaned and preprocessed to ensure that they meet the input requirements of the model. The preprocessed historical target spectrum of the transmission chain of each wind turbine and the root mean square index corresponding to the historical target spectrum are then used as training data to train the generative adversarial network model (such as AnoGAN). Specifically, adversarial training can be used to train the generator and discriminator. First, the parameters of the generator and discriminator (such as the weights and biases of the neural network) are initialized based on a random initialization method (such as a normal distribution or a uniform distribution). Next, the parameters of the generator are fixed, and the real data and the data generated by the generator through a random noise vector are input into the discriminator together. The discriminator is trained by minimizing the loss function of the discriminator so that it can better distinguish between real data and generated data. That is, the discriminator should be able to give a higher probability value to the real data and a lower probability value to the generated data. Then, the parameters of the discriminator are fixed, new data is generated through the generator, and it is input into the discriminator. According to the output of the discriminator (that is, the probability that the generated data is judged as real data), the parameters of the generator are adjusted so that the data generated by the generator can more effectively "cheat" the discriminator, that is, the probability that the discriminator judges the generated data as real data is higher. This process is achieved by minimizing the loss function of the generator. For example, the loss function used when training the discriminator can be a binary cross entropy loss function or a Among them, x represents the real data, represents the generated data, D(x) is the discriminator's judgment result on the real data, is the discriminant's judgment result on the generated data. loss_D(D(x),1) means that we hope that the discriminant's judgment result on the real data is as close as possible, that is, the discriminator can correctly judge the real data as real. It means that we hope that the discriminator can judge the generated data as close as possible, that is, the discriminator can correctly judge the generated data as false. Finally, based on the preset number of optimizations, the root mean square index corresponding to each historical target spectrum and the new loss function ( Among them, r is a vector representing input sample 1, f is another vector representing input sample 2, k is an integer representing the kth element of vector r or f, D(r) is the discrimination result of the discriminator with r as input, and D(f) is the discrimination result of the discriminator with f as input) The trained generative adversarial network model is optimized to obtain the transmission chain state evaluation model.
[0086] In addition, when using a new loss function to optimize the trained generative adversarial network model, the value obtained each time the loss function is calculated can be recorded, and then the preset fault threshold can be determined using these recorded values.
[0087] In this embodiment, the generative adversarial network model is trained by using the historical target spectrum of the transmission chain of each wind turbine and the root mean square index corresponding to the historical target spectrum, so that the obtained transmission chain state assessment model can be used to obtain the state assessment score of the transmission chain according to the target spectrum.
[0088] Step 216: Determine the fault diagnosis result of the transmission chain according to the state evaluation score.
[0089] Further, step 216 may specifically include: when it is determined that the state evaluation score is greater than a first threshold and the state evaluation score is not greater than a second threshold, determining that the fault diagnosis result is the occurrence of a first fault; when it is determined that the state evaluation score is greater than the second threshold, determining that the fault diagnosis result is the occurrence of a second fault; the first threshold is less than the second threshold.
[0090] Among them, the first threshold refers to a numerical standard for judging the fault state of the transmission chain. The second threshold refers to another numerical standard for judging the fault state of the transmission chain. For example, the first threshold can be 2 times the preset fault threshold, and the second threshold can be 3 times the preset fault threshold. The first fault refers to a lighter fault type determined according to the state evaluation score in fault diagnosis. For example, the first fault can be slight wear of the bearing. The second fault refers to a heavier fault type determined according to the state evaluation score in fault diagnosis, for example, the second fault can be serious damage to the bearing.
[0091] Specifically, after obtaining the status assessment score, it can be determined whether the status assessment score is greater than the first threshold value. If not, it is determined that the fault diagnosis result is that the transmission chain of the wind turbine set has not failed, and no further measures are required at this time. If it is greater than the first threshold value but not greater than the second threshold value, it is determined that the fault diagnosis result is that the transmission chain of the wind turbine set has a first fault. At this time, an early warning warning can be sent to the terminal of the staff to remind the staff to check the wind turbine set so as to promptly discover and deal with potential problems. If it is greater than the second threshold value, it is determined that the fault diagnosis result is that the transmission chain of the wind turbine set has a second fault, indicating that the fault is more serious or urgent. At this time, in addition to sending an alarm warning to the terminal of the staff, the staff should also be reminded to shut down the wind turbine set to prevent the fault from further deteriorating or causing danger. In addition, a warning and alarm triggering logic for continuous cumulative over-limit can also be designed, for example: when the cumulative excess of the first threshold value but not greater than the second threshold value reaches 3 times, the early warning mechanism is triggered, and when the cumulative excess of the second threshold value reaches 3 times, the alarm mechanism is triggered.
[0092] On the other hand, the first threshold and the second threshold can be determined by using the mean and standard deviation of all loss function values recorded when optimizing the model. The specific calculation formula is as follows: First threshold = (mean + 2 × standard deviation) × 10 第一调节参数 / 20 , the second threshold = (mean + 3 × standard deviation) × 10 第二调节参数 / 20 The first adjustment parameter and the second adjustment parameter are determined according to actual conditions or requirements.
[0093] In this embodiment, through the above steps, accurate classification of the fault degree can be achieved, which can help staff to formulate more targeted maintenance strategies to avoid excessive or insufficient maintenance, thereby ensuring the overall operating efficiency and reliability of the wind farm.
[0094] Therefore, the technical solution of the present invention, after dividing the sampling period into multiple sub-periods according to the sampling moment of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, determines the resampling moment of the vibration signal within each sub-period according to the minimum speed corresponding to the speed pulse signal within the sampling period, avoids the situation that the collected data is either too sparse to reflect key information or too dense to cause data redundancy due to the fixed sampling frequency, ensures that the signal can be processed at the optimal sampling rate in each sub-period, thereby improving the validity and accuracy of the data. After determining the vibration signal corresponding to each resampling moment, the spectrum of the vibration signal is determined according to the vibration signal corresponding to each resampling moment, which not only reduces the amount of data to be processed subsequently, but also improves the accuracy of the determined spectrum, thereby improving the accuracy of the subsequent fault diagnosis. Then, the spectrum is normalized according to the ratio of the speed corresponding to each frequency in the spectrum to the rated speed of the generator shaft, and the normalized spectrum of the vibration signal is obtained, which helps to improve the accuracy of the target spectrum determined subsequently. After that, the target spectrum is determined according to the normalized spectrum and the sampling moment of the speed pulse signal, which can better integrate the vibration characteristics and the speed characteristics, so that the accuracy of the transmission chain fault diagnosis result determined subsequently is higher. Then, the target spectrum is input into the pre-trained transmission chain state assessment model to obtain the state assessment score of the transmission chain, which can quickly obtain the state assessment score and improve the efficiency of subsequent diagnosis. Finally, the fault diagnosis result of the transmission chain is determined according to the state assessment score, which can accurately determine whether the transmission chain has a fault.
[0095] In addition, the resampling scale factor is determined according to the minimum speed, and then the resampling time of the sub-period is determined according to the resampling scale factor, which can convert the original vibration signal into an angular domain signal that is more suitable for subsequent analysis, which helps to simplify the signal processing process and reduce unnecessary calculation steps and data processing time. After that, the amplitude corresponding to the frequency point of the target spectrum is determined according to the adjacent frequency difference of the normalized spectrum, the index corresponding to the frequency point of the target spectrum, the interval between each sampling time of the speed pulse signal, and the amplitude corresponding to the frequency point of the normalized spectrum, which improves the accuracy of the determined target spectrum, thereby significantly improving the accuracy and reliability of subsequent fault diagnosis.
[0096] On the other hand, an adversarial network model is generated through the historical target spectrum of the transmission chain of each wind turbine and the root mean square indicator training data corresponding to the historical target spectrum, so that the obtained transmission chain state evaluation model can be used to obtain the state evaluation score of the transmission chain according to the target spectrum. Afterwards, when it is determined that the state evaluation score is greater than the first threshold and the state evaluation score is not greater than the second threshold, the fault diagnosis result is determined to be the occurrence of the first fault. When it is determined that the state evaluation score is greater than the second threshold, the fault diagnosis result is determined to be the occurrence of the second fault, which realizes the accurate classification of the degree of fault, thereby helping the staff to formulate more targeted maintenance strategies, avoid excessive or insufficient maintenance, and thus ensure the overall operation efficiency and reliability of the wind turbine.
[0097] Figure 3 A schematic structural diagram of a fault diagnosis device for a transmission chain of a wind turbine set provided in an embodiment of the present invention. The device and the fault diagnosis method for the transmission chain of a wind turbine set in the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiments of the fault diagnosis device for the transmission chain of a wind turbine set, reference can be made to the embodiments of the fault diagnosis method for the transmission chain of the wind turbine set in the above-mentioned embodiments.
[0098] like Figure 3 As shown, the device comprises:
[0099] A resampling module 310, configured to divide the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, and determine the resampling time of the vibration signal of the generator shaft within each of the sub-periods according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain;
[0100] A vibration spectrum determination module 320, configured to determine the vibration signal corresponding to each resampling moment and then determine the spectrum of the vibration signal according to the vibration signal corresponding to each resampling moment;
[0101] An evaluation score determination module 330 is used to determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain;
[0102] The diagnosis module 340 is used to determine the fault diagnosis result of the transmission chain according to the state evaluation score.
[0103] On the basis of the above embodiment, determining the resampling time of the vibration signal of the generator shaft in each sub-period according to the rotation speed of the rotation speed pulse signal includes:
[0104] The resampling time of the vibration signal in each sub-period is determined according to the lowest speed corresponding to the speed pulse signal in the sampling period.
[0105] On the basis of the above embodiment, determining the resampling time of the vibration signal in each sub-period according to the lowest speed corresponding to the speed pulse signal in the sampling period includes:
[0106] Determining a resampling scale factor according to the minimum rotation speed;
[0107] For a current sub-period, determining the number of samples of the current sub-period according to the duration of the current sub-period and the resampling scale factor;
[0108] The duration of the current sub-period is evenly divided based on the sampling quantity to obtain a resampling moment of the current sub-period.
[0109] On the basis of the above embodiment, determining the target spectrum according to the spectrum and the sampling time of the speed pulse signal includes:
[0110] Normalizing the spectrum according to the ratio of the rotation speed corresponding to each frequency in the spectrum to the rated rotation speed of the generator shaft to obtain a normalized spectrum of the vibration signal;
[0111] The target spectrum is determined according to the normalized spectrum and the sampling time of the rotation speed pulse signal.
[0112] On the basis of the above embodiment, the target spectrum is determined according to the normalized spectrum and the sampling time of the speed pulse signal, including:
[0113] Determining the frequency points of the normalized spectrum as the frequency points of the target spectrum;
[0114] Determining a frequency difference multiple according to adjacent frequency differences of the normalized spectrum and indexes corresponding to frequency points of the target spectrum;
[0115] Determine the amplitude corresponding to the frequency point of the target spectrum according to the frequency difference multiple, the interval between each sampling moment of the speed pulse signal and the amplitude corresponding to the frequency point of the normalized spectrum;
[0116] The target spectrum is constructed according to the frequency points of the target spectrum and the amplitudes corresponding to the frequency points of the target spectrum.
[0117] Based on the above embodiment, the training process of the transmission chain state assessment model includes:
[0118] Acquire a historical target frequency spectrum of a transmission chain of each of the wind turbine generator sets and a root mean square index corresponding to the historical target frequency spectrum;
[0119] The historical target frequency spectrum of the transmission chain of each wind turbine generator set and the root mean square index corresponding to the historical target frequency spectrum are used as training data to perform model training on the generative adversarial network model to obtain the transmission chain state evaluation model.
[0120] Based on the above embodiment, the diagnosis module 340 is specifically used for:
[0121] When it is determined that the state evaluation score is greater than a first threshold value and the state evaluation score is not greater than a second threshold value, determining that the fault diagnosis result is that a first fault occurs;
[0122] When it is determined that the state evaluation score is greater than the second threshold, the fault diagnosis result is determined to be the occurrence of a second fault; and the first threshold is less than the second threshold.
[0123] The fault diagnosis device for the transmission chain of a wind turbine set provided in the embodiment of the present invention can execute the fault diagnosis method for the transmission chain of a wind turbine set provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0124] It is worth noting that in the embodiment of the fault diagnosis device for the transmission chain of the above-mentioned wind turbine generator set, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0125] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 4 A block diagram of an exemplary electronic device 4 suitable for implementing embodiments of the present invention is shown. Figure 4 The electronic device 4 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0126] like Figure 4 As shown, the electronic device 4 is in the form of a general purpose computing electronic device. The components of the electronic device 4 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).
[0127] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.
[0128] The electronic device 4 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 4, including volatile and non-volatile media, removable and non-removable media.
[0129] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 4 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.
[0130] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.
[0131] The electronic device 4 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the electronic device 4, and / or any device that enables the electronic device 4 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the electronic device 4 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. Figure 4 As shown, the network adapter 20 communicates with other modules of the electronic device 4 via the bus 18. It should be understood that although Figure 4 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 4, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0132] The processing unit 16 executes various functional applications and page displays by running the program stored in the system memory 28, for example, implementing the fault diagnosis method of the transmission chain of the wind turbine provided by the embodiment of the present invention, the method comprising:
[0133] After dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, the resampling time of the vibration signal of the generator shaft within each of the sub-periods is determined according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain;
[0134] After determining the vibration signal corresponding to each of the resampling moments, determining the frequency spectrum of the vibration signal according to the vibration signal corresponding to each of the resampling moments;
[0135] Determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain;
[0136] A fault diagnosis result of the transmission chain is determined according to the state evaluation score.
[0137] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the fault diagnosis method of the transmission chain of the wind turbine set provided by any embodiment of the present invention.
[0138] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, a fault diagnosis method for a transmission chain of a wind turbine generator set provided by an embodiment of the present invention is implemented. The method includes:
[0139] After dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, the resampling time of the vibration signal of the generator shaft within each of the sub-periods is determined according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain;
[0140] After determining the vibration signal corresponding to each of the resampling moments, determining the frequency spectrum of the vibration signal according to the vibration signal corresponding to each of the resampling moments;
[0141] Determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain;
[0142] A fault diagnosis result of the transmission chain is determined according to the state evaluation score.
[0143] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.
[0144] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0145] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0146] Computer program code for performing the operations of the present invention may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0147] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
[0148] In addition, the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.
[0149] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims
1. A fault diagnosis method for a transmission chain of a wind turbine generator set, characterized in that: The method comprises: After dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, the resampling time of the vibration signal of the generator shaft within each of the sub-periods is determined according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain; After determining the vibration signal corresponding to each of the resampling moments, determining the frequency spectrum of the vibration signal according to the vibration signal corresponding to each of the resampling moments; Determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain; A fault diagnosis result of the transmission chain is determined according to the state evaluation score.
2. The fault diagnosis method of the transmission chain of a wind turbine generator set according to claim 1, characterized in that: Determining the resampling time of the vibration signal of the generator shaft in each sub-period according to the rotation speed of the rotation speed pulse signal includes: The resampling time of the vibration signal in each sub-period is determined according to the lowest speed corresponding to the speed pulse signal in the sampling period.
3. The fault diagnosis method for the transmission chain of a wind turbine generator set according to claim 2, characterized in that: Determining the resampling time of the vibration signal in each sub-period according to the lowest speed corresponding to the speed pulse signal in the sampling period includes: Determining a resampling scale factor according to the minimum rotation speed; For a current sub-period, determining the number of samples of the current sub-period according to the duration of the current sub-period and the resampling scale factor; The duration of the current sub-period is evenly divided based on the sampling quantity to obtain a resampling moment of the current sub-period.
4. The fault diagnosis method for the transmission chain of a wind turbine generator set according to claim 1, characterized in that: Determining a target spectrum according to the spectrum and the sampling time of the speed pulse signal includes: Normalizing the spectrum according to the ratio of the rotation speed corresponding to each frequency in the spectrum to the rated rotation speed of the generator shaft to obtain a normalized spectrum of the vibration signal; The target spectrum is determined according to the normalized spectrum and the sampling time of the rotation speed pulse signal.
5. The fault diagnosis method of the transmission chain of a wind turbine generator set according to claim 4, characterized in that: Determining the target spectrum according to the normalized spectrum and the sampling time of the speed pulse signal includes: Determining the frequency points of the normalized spectrum as the frequency points of the target spectrum; Determining a frequency difference multiple according to adjacent frequency differences of the normalized spectrum and indexes corresponding to frequency points of the target spectrum; Determine the amplitude corresponding to the frequency point of the target spectrum according to the frequency difference multiple, the interval between each sampling moment of the speed pulse signal and the amplitude corresponding to the frequency point of the normalized spectrum; The target spectrum is constructed according to the frequency points of the target spectrum and the amplitudes corresponding to the frequency points of the target spectrum.
6. The fault diagnosis method of the transmission chain of a wind turbine generator set according to claim 1, characterized in that: The training process of the transmission chain state assessment model includes: Acquire a historical target frequency spectrum of a transmission chain of each of the wind turbine generator sets and a root mean square index corresponding to the historical target frequency spectrum; The historical target frequency spectrum of the transmission chain of each wind turbine generator set and the root mean square index corresponding to the historical target frequency spectrum are used as training data to perform model training on the generative adversarial network model to obtain the transmission chain state evaluation model.
7. The fault diagnosis method of the transmission chain of a wind turbine generator set according to claim 1, characterized in that: Determining a fault diagnosis result of the transmission chain according to the state evaluation score includes: When it is determined that the state evaluation score is greater than a first threshold value and the state evaluation score is not greater than a second threshold value, determining that the fault diagnosis result is that a first fault occurs; When it is determined that the state evaluation score is greater than the second threshold, the fault diagnosis result is determined to be the occurrence of a second fault; and the first threshold is less than the second threshold.
8. A fault diagnosis device for a transmission chain of a wind turbine generator set, characterized in that: The device comprises: A resampling module, for dividing the sampling period into a plurality of sub-periods according to the sampling time of the speed pulse signal of the generator shaft of the wind turbine generator set within the sampling period, and determining the resampling time of the vibration signal of the generator shaft within each sub-period according to the speed of the speed pulse signal, wherein the wind turbine generator set includes a generator shaft and a transmission chain; A vibration spectrum determination module, configured to determine the spectrum of the vibration signal according to the vibration signal corresponding to each resampling moment after determining the vibration signal corresponding to each resampling moment; An evaluation score determination module is used to determine a target spectrum according to the spectrum and the sampling time of the speed pulse signal, and input the target spectrum into a pre-trained transmission chain state evaluation model to obtain a state evaluation score of the transmission chain; A diagnosis module is used to determine a fault diagnosis result of the transmission chain according to the state evaluation score.
9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the fault diagnosis method for the transmission chain of a wind turbine set as described in any one of claims 1-7.
10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, implement the fault diagnosis method for the transmission chain of a wind turbine set as claimed in any one of claims 1 to 7.