Method, device, equipment, storage medium and product for detecting failure of wind turbine generator system

Abstract

The application discloses a wind turbine fault detection method, device, equipment, storage medium and product. The wind turbine fault detection method comprises the following steps: acquiring first operation data of a wind turbine and operation parameter data set of the wind turbine in a preset time period, wherein the wind turbine comprises multiple operation stages, and the operation parameter data set comprises multiple second operation data corresponding to each operation stage; determining membership of the first operation data corresponding to each operation stage according to the first operation data and the multiple second operation data corresponding to each operation stage; determining that an operation stage corresponding to membership satisfying a preset condition is a target operation stage corresponding to the first operation data; and determining a fault detection result of the first operation data according to a preset fault analysis model corresponding to the target operation stage. According to the embodiment of the application, the problem that the accuracy of the fault detection of the wind turbine is low can be solved.

Description

Technical Field

[0001] This application belongs to the field of wind power generation technology, and in particular relates to a method, apparatus, equipment, storage medium and product for fault detection of wind turbine generator sets. Background Technology

[0002] Wind power generation, as a green and renewable energy source, has been increasingly widely adopted. However, with the increase in the number of wind turbines and their accumulated operating time, the probability of wind turbine failures also increases relatively.

[0003] A failure in a wind turbine generator set can lead to a decline in its performance. To ensure the normal operation of wind turbine generator sets, it is often necessary to collect operational data and use this data for fault detection, so as to adjust the operation and maintenance strategies in a timely manner. However, the inventors of this application have discovered that in related technologies, fault detection is performed directly based on a preset analysis model using the collected operational data, which results in low accuracy in fault detection of wind turbine generator sets. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and product for fault detection of wind turbine generator sets, which can solve the problem of low accuracy in fault detection of wind turbine generator sets.

[0005] In a first aspect, embodiments of this application provide a method for fault detection of a wind turbine generator set, including:

[0006] Acquire the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period. The wind turbine generator set includes multiple operating stages, and the operating parameter dataset includes multiple second operating data corresponding to each operating stage.

[0007] Based on the first running data and multiple second running data corresponding to each running stage, determine the membership degree of the first running data to each running stage;

[0008] The operational stage corresponding to the membership degree that meets the preset conditions is determined as the target operational stage corresponding to the first operational data;

[0009] Based on the preset fault analysis model corresponding to the target operation stage, the fault detection results of the first operation data are determined.

[0010] In some possible implementations of the first aspect, the membership degree of the first running data to each running stage is determined based on the first running data and multiple second running data corresponding to each running stage, including:

[0011] Based on the first running data and multiple second running data corresponding to each running stage, determine the Mahalanobis distance of the first running data for each running stage;

[0012] Based on the first running data and multiple Mahalanobis distances, the membership degree between the first running data and each running stage is determined.

[0013] In some possible implementations of the first aspect, the Mahalanobis distance for each running stage is determined based on the first running data and multiple second running data corresponding to each running stage, including:

[0014] Based on multiple second-stage data corresponding to each stage of operation, determine the covariance matrix and mean for each stage of operation.

[0015] Based on the first set of running data and the covariance matrix and mean of each running stage, determine the Mahalanobis distance for each running stage corresponding to the first set of running data.

[0016] In some possible implementations of the first aspect, the membership degree between the first running data and each running stage is determined based on the first running data and multiple Mahalanobis distances, including:

[0017] Based on multiple Mahalanobis distance and membership degree calculation formulas, the membership degree of the first running data for each running stage is determined. The membership degree calculation formula is as follows:

[0018]

[0019] in, The Mahalanobis distance for each running stage corresponds to the first running data. For the first running data corresponding to the first i Membership degree of each operational phase The Mahalanobis distance for each running stage corresponds to the first running data. M The total number of operation phases. M positive integer, m As a preset constant, m >1.

[0020] In some possible implementations of the first aspect, determining the operational stage corresponding to the membership degree that satisfies preset conditions as the target operational stage corresponding to the first operational data includes:

[0021] The operational stage corresponding to the largest membership degree among multiple membership degrees is determined as the target operational stage corresponding to the first operational data.

[0022] In some possible implementations of the first aspect, multiple operational phases include two adjacent operational phases; determining the operational phase corresponding to the membership degree that satisfies preset conditions, as the target operational phase corresponding to the first operational data, includes:

[0023] Calculate the harmonic mean of the membership degrees of two adjacent running stages in a series of running stages;

[0024] If the harmonic average meets the preset threshold range, the target operating phase is determined to include two adjacent operating phases.

[0025] In some possible implementations of the first aspect, the fault detection results of the first operational data are determined based on a preset fault analysis model corresponding to the target operational phase, including:

[0026] Based on the preset fault analysis model corresponding to any one of the two adjacent operation phases, determine the fault detection result of the first operation data.

[0027] In some possible implementations of the first aspect, before acquiring the first operating data of the wind turbine generator and the dataset of operating parameters of the wind turbine generator within a preset time period, the method further includes:

[0028] Acquire multiple third-party operating data of the wind turbine generator set within a preset time period, wherein each third-party operating data includes the wind turbine generator set's rotational speed, torque, and pitch angle.

[0029] Pre-set statistical analysis is performed on multiple speeds, torques, and pitch angles of the wind turbine generator set to determine the operating stage corresponding to each third operating data point, and multiple second operating data points corresponding to each operating stage are obtained.

[0030] In some possible implementations of the first aspect, multiple third-party operating data of the wind turbine generator within a preset time period are obtained, including:

[0031] Third operating data that does not meet the preset operating conditions of the wind turbine generator set are filtered out from multiple third operating data of the wind turbine generator set.

[0032] In some feasible implementations of the first aspect, multiple operating phases include a startup phase and an optimal wind energy tracking phase; pre-defined statistical analysis is performed on multiple speeds, torques, and pitch angles of the wind turbine generator to determine the operating phase corresponding to each third operating data point, including:

[0033] According to the preset speed division rules, the multiple speeds of the wind turbine generator are divided to obtain multiple speed compartments;

[0034] Determine the mean torque in each speed chamber, the dispersion of torque in each speed chamber, and determine the first fitting curve of the mean torque and the corresponding speed of each speed chamber, and the second fitting curve of the dispersion of torque and the corresponding speed of each speed chamber.

[0035] Based on the tangent change rate of the first fitted curve and the tangent change rate of the second fitted curve, the operating stage corresponding to each third operating data point is determined to be either the startup stage or the optimal wind energy tracking stage.

[0036] In some feasible implementations of the first aspect, the multiple operating stages also include a rated speed stage and a rated power stage, and the third operating data also includes multiple operating power levels of the wind turbine generator set; a pre-defined statistical analysis is performed on multiple speeds, multiple torques, and multiple pitch angles of the wind turbine generator set to determine the operating stage corresponding to each third operating data point, including:

[0037] According to the preset power division rules, the multiple operating powers of the wind turbine generator set are divided to obtain multiple power compartments;

[0038] Determine the maximum pitch angle in each power compartment, and determine the third fitting curve between the maximum pitch angle and the corresponding power of each power compartment;

[0039] Based on the rate of change of the tangent of the third fitted curve, the operating stage corresponding to each third operating data point is determined to be either the rated speed stage or the rated power stage.

[0040] Secondly, embodiments of this application provide a device for detecting faults in wind turbine generator sets, comprising:

[0041] The acquisition module is used to acquire the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period. The wind turbine generator set includes multiple operating stages, and the operating parameter dataset includes multiple second operating data corresponding to each operating stage.

[0042] The processing module is used to determine the membership degree of the first running data to each running stage based on the first running data and multiple second running data corresponding to each running stage;

[0043] The processing module is also used to determine the running stage corresponding to the membership degree that meets the preset conditions, which is the target running stage corresponding to the first running data;

[0044] The processing module is also used to determine the fault detection results of the first operating data based on the preset fault analysis model corresponding to the target operating stage.

[0045] Thirdly, this application provides a device for detecting faults in wind turbine generator sets, the device comprising: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the method for detecting faults in wind turbine generator sets as described in the first aspect or any implementable embodiment of the first aspect.

[0046] Fourthly, this application provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the method for wind turbine generator fault detection as described in the first aspect or any implementable method of the first aspect.

[0047] Fifthly, this application provides a computer program product in which the instructions are executed by the processor of an electronic device, causing the electronic device to perform the method for detecting wind turbine generator set faults as described in the first aspect or any implementable method of the first aspect.

[0048] This application provides a method, apparatus, device, and readable storage medium for fault detection of wind turbine generator sets. In this application embodiment, firstly, first operating data and a dataset of operating parameters of the wind turbine generator set within a preset time period are acquired. Since the wind turbine generator set includes multiple operating stages, the dataset of operating parameters includes multiple second operating data corresponding to each operating stage. Next, based on the first operating data and the multiple second operating data corresponding to each operating stage, the membership degree of the first operating data to each operating stage is determined; and the operating stage corresponding to the membership degree that meets preset conditions is determined as the target operating stage corresponding to the first operating data. Then, based on a preset fault analysis model corresponding to the target operating stage, the fault detection result of the first operating data is determined. Because the operating data is divided into stages, a fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, thereby effectively improving the detection accuracy of faults in wind turbine generator sets. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a method for fault detection of a wind turbine generator set provided in an embodiment of this application;

[0051] Figure 2 This is a schematic diagram of the division of operation stages provided in an embodiment of this application;

[0052] Figure 3 This is another schematic diagram of the division of operation stages provided in the embodiments of this application;

[0053] Figure 4 This is a schematic diagram of multiple running stages corresponding to a running parameter dataset provided in an embodiment of this application;

[0054] Figure 5 This is a schematic diagram of the structure of a wind turbine generator fault detection device provided in an embodiment of this application;

[0055] Figure 6 This is a schematic diagram of the structure of a wind turbine generator set fault detection device provided in an embodiment of this application. Detailed Implementation

[0056] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended only to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0057] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0058] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0059] Wind power generation, as a green and renewable energy source, has been increasingly widely adopted. However, with the increase in the number of wind turbines and their accumulated operating time, the probability of wind turbine failures also increases relatively.

[0060] A failure in a wind turbine generator set can lead to a decline in its performance. To ensure the normal operation of wind turbine generator sets, it is often necessary to collect operational data and combine this data for fault detection, so as to adjust the operation and maintenance strategies of the wind turbine generator set in a timely manner. However, current commonly used fault detection methods, such as directly analyzing and modeling the operational data collected from the wind turbine generator set, or performing cluster analysis solely based on the similarity of environmental data, have revealed that current fault detection methods for wind turbine generator sets do not fully utilize control logic experience and suffer from insufficient feature recognition of operational data, resulting in low accuracy in fault detection.

[0061] In view of the inventors' above-mentioned research findings, this application provides a method, apparatus, device, storage medium, and product for wind turbine generator set fault detection, which can solve the problem of low accuracy in fault detection of wind turbine generator sets. The technical concept of this application is as follows: by acquiring first operating data of the wind turbine generator set and a dataset of operating parameters of the wind turbine generator set within a preset time period, the membership degree of each operating stage corresponding to the first operating data can be determined based on the first operating data and the operating parameter dataset; and the operating stage corresponding to the membership degree that meets preset conditions is determined as the target operating stage corresponding to the first operating data. Then, based on a preset fault analysis model corresponding to the target operating stage, the fault detection result of the first operating data is determined.

[0062] The method for fault detection of wind turbine generator sets provided in the embodiments of this application is described below with reference to the accompanying drawings.

[0063] Figure 1 A flowchart illustrating a method for fault detection of a wind turbine generator set according to an embodiment of this application is shown. Figure 1 As shown, the method may include steps 110 to 140.

[0064] Step 110: Obtain the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period.

[0065] The wind turbine generator set includes multiple operating phases, and the operating parameter dataset includes multiple second operating data corresponding to each operating phase.

[0066] Step 120: Determine the membership degree of the first running data for each running stage based on the first running data and the multiple second running data corresponding to each running stage.

[0067] Step 130: Determine the running stage corresponding to the membership degree that meets the preset conditions, which is the target running stage corresponding to the first running data.

[0068] Step 140: Determine the fault detection results of the first operating data based on the preset fault analysis model corresponding to the target operating stage.

[0069] The specific implementation methods of the above steps will be described in detail below.

[0070] In this embodiment, firstly, first operating data of the wind turbine generator set and operating parameter dataset of the wind turbine generator set within a preset time period are acquired. Since the wind turbine generator set includes multiple operating stages, the operating parameter dataset includes multiple second operating data corresponding to each operating stage. Next, based on the first operating data and the multiple second operating data corresponding to each operating stage, the membership degree of the first operating data to each operating stage can be determined; and the operating stage corresponding to the membership degree that meets the preset conditions is determined as the target operating stage corresponding to the first operating data. In this way, the control logic experience of the wind turbine generator set in different operating stages can be combined to determine the target operating stage corresponding to the first operating data. Next, based on the preset fault analysis model corresponding to the target operating stage, the fault detection result of the first operating data is determined. Since the operating data is divided into stages, the fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, thereby effectively improving the detection accuracy of fault detection of the wind turbine generator set.

[0071] The specific implementation methods for each of the above steps are described below.

[0072] Step 110: Obtain the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period.

[0073] First, in step 110, the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period are obtained.

[0074] In this embodiment, the first operating data of the wind turbine generator set may include real-time operating data of the wind turbine generator set, such as the wind turbine's rotational speed, torque, pitch angle, operating power, etc. This real-time operating data can be acquired in real time by a data acquisition unit installed on the wind turbine generator set.

[0075] The preset time period can be a period of time prior to the acquisition of the first operational data. The length of the preset time period can be several hours, several days, etc., and can be set according to actual application needs; no specific limitation is made here. The operational data generated by the wind turbine generator within the preset time period can be stored in a preset database. For example, a Supervisory Control and Data Acquisition (SCADA) system can be used to store the wind turbine generator's operational data.

[0076] A wind turbine generator set may include multiple operating phases. For example, the operating phases of a wind turbine generator set may include a startup phase, an optimal wind energy tracking phase, a rated speed phase, and a rated power phase. In this embodiment, second operating data corresponding to any one of the above operating phases can be obtained from a preset database.

[0077] The second operating data can be the operating data generated by the wind turbine generator within a preset time period, such as the wind turbine's speed, torque, pitch angle, and operating power within that time period. For each piece of second operating data stored in a preset database, each piece of data is labeled with the corresponding operating status information of the wind turbine generator. For example, the operating status information of the wind turbine generator may include information on the wind turbine generator's operating stage, power limitation information, and operating fault information. Different operating stages of the wind turbine generator can correspond to different operating states of the wind turbine generator, which are not listed here.

[0078] According to the embodiments of this application, by extracting the corresponding second operating data for each operating stage, the operating parameter dataset of the wind turbine generator set within a preset time period can be obtained.

[0079] In this embodiment, the wind turbine generating the first operating data and the wind turbine corresponding to the operating parameter dataset can be the same wind turbine. Optionally, when the amount of data including the second operating data in the operating parameter dataset cannot meet the analysis requirements, other wind turbines with the same or similar operating environment as the wind turbine generating the first operating data and using the same or similar control strategies can be selected according to actual needs, and the operating data generated by such wind turbines can be included in the operating parameter dataset. No specific limitations are imposed here.

[0080] In some embodiments, in order to accurately determine the operating status information of the wind turbine generator corresponding to the operating data generated by the wind turbine generator within a preset time period, so as to make full use of the data characteristics of different operating stages, optionally, the operating status information of the wind turbine generator corresponding to the operating data can be determined according to steps 210 to 220, thereby obtaining the operating stage corresponding to the operating data. For the sake of brevity and clarity in describing the embodiments of this application, before determining the operating status information of the wind turbine generator corresponding to the operating data, the operating data generated by the wind turbine generator is described as third operating data.

[0081] Step 210: Obtain multiple third-party operating data of the wind turbine generator set within a preset time period.

[0082] Step 220: Perform preset statistical analysis on multiple speeds, multiple torques, and multiple pitch angles of the wind turbine generator set to determine the operating stage corresponding to each third operating data point, and obtain multiple second operating data points corresponding to each operating stage.

[0083] Specifically, each third operating data point may include the wind turbine's rotational speed, torque, and pitch angle.

[0084] To improve the reliability of the operating parameter dataset, optionally, the following operation can be performed on the third operating data within a preset time period: filtering out third operating data that does not meet preset operating conditions from multiple third operating data sets of the wind turbine generator set. The preset operating conditions can be set based on at least one of the following aspects, such as: whether the wind turbine generator set has a fault, whether the wind turbine generator set is operating with limited power, whether the wind turbine generator set is in grid-connected power generation, etc., without specific limitations. Thus, based on the preset operating conditions, operating data corresponding to a healthy and stable state of the wind turbine generator set can be obtained, thereby improving the reliability of the extracted operating parameter dataset.

[0085] In some embodiments, each third running data point can be collected according to a preset data collection time interval. For example, each third running data point can be considered as a sample data point for each collection moment. Optionally, the number of sample data points can be denoted as... l The control variables of the wind turbine generator in the third operating data can be denoted as: d The control variables are operating data such as speed, torque, and pitch angle.

[0086] The system includes pre-defined statistical analyses, such as calculating the mean, variance, and coefficient of variation for the third set of running data. Based on the data characteristics of each running stage, it determines the running stage to which each set of third running data belongs, and obtains multiple sets of second running data corresponding to each running stage. Next, the second running data can be stored, and the running stages corresponding to the second running data can be marked. This allows for convenient and rapid extraction of the second running data when the running stage corresponding to the first running data is determined.

[0087] In some embodiments, to improve computation speed, after obtaining multiple second operational data points corresponding to each operational stage, statistical analysis can be directly performed based on the second operational data points corresponding to each operational stage to calculate the statistical parameters corresponding to each operational stage. Statistical parameters include, for example, the covariance matrix and the mean. If the statistical parameters corresponding to each operational stage are needed, they can be extracted directly without repeated calculations.

[0088] As a specific embodiment, the operation phase of a wind turbine generator set may include a startup phase and an optimal wind energy tracking phase. In order to accurately determine the operation phase corresponding to the third operation data, it can be determined according to steps 221 to 223.

[0089] Step 221: According to the preset speed division rules, the speed of multiple wind turbine generator sets is divided to obtain multiple speed compartments.

[0090] For example, the preset speed division rules can be, for instance, to set multiple speed ranges corresponding to different speed magnitudes. Specifically, for example, if the speed range corresponding to multiple third operating data is 10~30 rpm, the length of each speed range can be set to 0.5. After dividing the speed of multiple wind turbine generator sets, the resulting speed compartments can include, for example, [10, 10.5), [10.5, 11), [11, 11.5), ..., [29.5, 30].

[0091] Optionally, the length of each speed range can be set according to the actual data analysis needs, and is not specifically limited here.

[0092] Step 222: Determine the mean torque in each speed chamber, the dispersion of torque in each speed chamber, and determine the first fitting curve of the mean torque and the corresponding speed of each speed chamber, and the second fitting curve of the dispersion of torque and the corresponding speed of each speed chamber.

[0093] Specifically, since each third set of operational data can be collected according to a preset data acquisition time interval, corresponding operational data such as rotational speed, torque, and pitch angle can be obtained at each acquisition moment. Therefore, after dividing the data into sections based on rotational speed, each rotational speed section can include data such as torque and pitch angle. Thus, the average torque of each rotational speed section can be determined based on the torque included in each section.

[0094] The dispersion of torque can be represented using statistical indicators such as variance, coefficient of variation, or standard deviation. After dividing the torque into speed compartments, the dispersion of torque corresponding to each speed compartment can be determined based on the torque included in each speed compartment.

[0095] In some embodiments, the rotational speed corresponding to each rotational speed chamber can be the maximum, minimum, or average value of the rotational speed chamber, etc., without specific limitation. Thus, a first fitting curve can be obtained by fitting data based on the average torque corresponding to each rotational speed chamber and the corresponding rotational speed, and a second fitting curve can be obtained by fitting data based on the dispersion of the torque corresponding to each rotational speed chamber and the corresponding rotational speed. In the embodiments of this application, the data fitting method can, for example, use the least squares curve fitting method, etc., without specific limitation.

[0096] Step 223: Based on the tangent change rate of the first fitted curve and the tangent change rate of the second fitted curve, determine the operating stage corresponding to each third operating data as the startup stage or the optimal wind energy tracking stage.

[0097] Based on the fluctuations of the first and second fitted curves, the boundary points of different operating stages can be determined, and thus the operating stage corresponding to each third operating data point can be determined.

[0098] For example, Figure 2 This is a schematic diagram illustrating the division of operational phases provided in an embodiment of this application. Taking phase A as an example, phase B as the startup phase, phase C as the optimal wind energy tracking phase, and phase D as the rated speed phase, the diagram illustrates this. Combined with... Figure 2 As shown, the horizontal axis represents rotational speed, and the vertical axis represents torque. The first fitting curve is obtained by fitting the mean values ​​of rotational speed and torque, and the second fitting curve is obtained by fitting the dispersion of rotational speed and torque. The boundary point of stage AB can be determined based on the rate of change of the tangent of the first fitting curve, and the boundary point of stage BC can be determined based on the rate of change of the tangent of the second fitting curve. After determining the boundary points of stages AB and BC, we can obtain, for example... Figure 2 The four regions shown are (a), (b), (c), and (d). Figure 2 Region (b) corresponds to stage A. Figure 2 Region (c) corresponds to stage A. Figure 2Region (d) corresponds to stage A. Wherein, Figure 2 Region (a) corresponds to the operating stage of the operating data. Power analysis can be combined to improve the accuracy of determining the operating stage corresponding to the third operating data.

[0099] Thus, according to the operation stage division method provided in the embodiments of this application, by analyzing the statistical indicators of torque corresponding to different speed chambers, the data characteristics of different operation stages can be understood, thereby accurately determining the operation stage corresponding to the third operation data.

[0100] As another concrete example, after obtaining the torque dispersion corresponding to each speed range, taking the coefficient of variation as an example, after determining the coefficient of variation of torque in each speed range, the speed control type corresponding to each of the third operating data can be determined based on the position of the extreme points in the second fitted curve. For example, the speed control type may include conventional control type, speed skipping control type, and optimization control type, etc., which will not be listed here. (Continue to refer to...) Figure 2 The second fitted curve includes two maxima, thus confirming that the speed control type is a conventional control type. This allows for understanding the unit's control strategy mode from a data perspective. Comparing this with the predetermined unit control strategy not only helps optimize the control strategy but also allows for verification of whether the wind turbine is operating according to the predetermined unit control strategy.

[0101] In some embodiments, the multiple operating phases may further include a rated speed phase and a rated power phase, and the third operating data may further include the operating power of multiple wind turbine generator sets. In determining the operating phase corresponding to each third operating data, steps 301 to 303 may be specifically included.

[0102] Step 301: According to the preset power division rules, the operating power of multiple wind turbine generator sets is divided to obtain multiple power compartments.

[0103] For example, the preset power division rules could be, for instance, setting multiple power ranges corresponding to different power levels. Optionally, the length of each power range can be specifically set according to actual data analysis needs, and is not specifically limited here.

[0104] Step 302: Determine the maximum value of the pitch angle in each power compartment, and determine the third fitting curve of the maximum value of the pitch angle and the corresponding power of each power compartment.

[0105] Specifically, since each third operational data point can be collected according to a preset data acquisition time interval, corresponding operational data such as the operating power of the wind turbine generator can also be obtained at each acquisition time. Therefore, after power-based segmentation, each power segment can include data such as pitch angle. Thus, the maximum value of the pitch angle corresponding to each power segment can be determined based on the pitch angle included in each power segment.

[0106] In some embodiments, the power corresponding to each power cell can be the maximum, minimum, or average value of the power cells, etc., without specific limitations. Thus, a third fitting curve can be obtained by fitting data based on the maximum pitch angle corresponding to each power cell and the power corresponding to each power cell. In the embodiments of this application, the data fitting method can be, for example, the least squares curve fitting method, etc., without specific limitations.

[0107] Step 303: Based on the rate of change of the tangent of the third fitted curve, determine whether the operating stage corresponding to each third operating data point is the rated speed stage or the rated power stage.

[0108] Based on the fluctuation of the third fitted curve, the dividing point between the rated speed stage and the rated power stage can be determined, and thus the corresponding operating stage for each third operating data point can be determined.

[0109] Figure 3 This is another schematic diagram of the operation stage division provided in the embodiments of this application. The rated speed stage corresponds to stage C, and the rated power stage corresponds to stage D. Combined with... Figure 3 As shown, the horizontal axis represents power, and the vertical axis represents the pitch angle. The third fitting curve is obtained by fitting the maximum values ​​of power and pitch angle. Optionally, to improve the accuracy of the fitting function, all third-stage running data can be fitted to obtain a third fitting curve. Based on the rate of change of the tangent of the third fitting curve, the boundary points of the CD stage on the horizontal and vertical axes can be determined respectively. Thus, for example, we can obtain... Figure 3 The four regions shown are (a), (b), (c), and (d). Figure 3 Region (c) corresponds to stage C. Figure 3 Regions (a), (b), and (d) correspond to stage D. Thus, according to the operation stage division method provided in the embodiments of this application, by analyzing the statistical indicators of the pitch angle corresponding to different power chambers, the data characteristics of different operation stages can be understood, thereby accurately determining the operation stage corresponding to the third operation data.

[0110] Continue to combine Figure 3As shown, if the third operating data in the third fitted curve has already yielded the corresponding operating stage based on the speed analysis, then it is only necessary to determine the third operating data for which the corresponding operating stage has not been obtained based on the power analysis. Based on the above analysis of the third operating data, the accuracy of determining the operating stage corresponding to the third operating data can be improved.

[0111] To provide a concise and clear description of the embodiments of this application, the third operating data that determines the operating status information of the wind turbine generator set will be described as the second operating data.

[0112] Optionally, when storing the second operating data, the corresponding operating status information of the wind turbine generator can be marked for each second operating data. Therefore, the operating status information of the wind turbine generator includes information on the operating stage of the wind turbine generator. Optionally, the statistical parameters corresponding to each operating stage can also be directly saved, without specific limitations.

[0113] According to the embodiments of this application, the operating parameter dataset of the wind turbine generator set within a preset time period can be quickly extracted from a preset database to facilitate the determination of the operating stage to which the first operating data of the wind turbine generator set belongs.

[0114] In step 120, the membership degree of the first running data corresponding to each running stage is determined based on the first running data and multiple second running data corresponding to each running stage.

[0115] Membership degree can be used to represent the probability that the first running data belongs to a certain running stage. The higher the membership degree calculated for a certain running stage, the greater the probability that the first running data belongs to that running stage.

[0116] Optionally, a preset membership function can be used to calculate the first running data and multiple second running data corresponding to each running stage. Optionally, the method for determining the preset membership function can be, for example, fuzzy statistical method, subjective experience method, etc., and is not specifically limited here.

[0117] In this embodiment of the application, in order to improve the reliability of the membership calculation results, specifically, the membership degree between the first running data and each running stage is determined, which can be based on the following steps:

[0118] First, based on the first running data and multiple second running data corresponding to each running stage, determine the Mahalanobis distance between the first running data and each running stage; next, based on the first running data and multiple Mahalanobis distances, determine the membership degree between the first running data and each running stage.

[0119] According to an embodiment of this application, Mahalanobis distance can be used to represent the distance from a sample point to the center point of the variable required for dividing each running stage. The smaller the distance, the greater the probability that the sample point belongs to the running stage corresponding to the minimum distance. For example, the sample point can be the first running data, and the center point of the variable can be determined based on multiple second running data corresponding to each running stage.

[0120] For example, Figure 4 This is a schematic diagram illustrating multiple running stages corresponding to a running parameter dataset provided in an embodiment of this application. Combined with... Figure 4 As shown, the first operating data includes the control variables of the wind turbine generator set, namely operating data such as speed, torque, and pitch angle. The wind turbine generator set includes multiple operating stages; for example, the startup stage corresponds to stage A, the optimal wind energy tracking stage corresponds to stage B, the rated speed stage corresponds to stage C, and the rated power stage corresponds to stage D. Multiple second operating parameters corresponding to each operating stage can be referenced. Figure 4 As shown.

[0121] In some embodiments, the calculation method for the Mahalanobis distance of the first running data corresponding to each running stage is as follows: based on multiple second running data corresponding to each running stage, determine the covariance matrix and mean of each running stage; based on the first running data and the covariance matrix and mean of each running stage, determine the Mahalanobis distance of the first running data corresponding to each running stage.

[0122] For example, the Mahalanobis distance corresponding to each running stage of the first running data can be calculated according to formula (1). Alternatively, for ease of calculation, the square of the Mahalanobis distance can be used. .

[0123] (1)

[0124] in, For the first running data subvector, For the first i The mean of the second set of running data during the running phase. For the first i The inverse of the covariance matrix of the second running data during the running phase.

[0125] Since there are different measurement standards for operating data such as speed, torque, and pitch angle, by calculating the Mahalanobis distance corresponding to each operating stage of the first operating data, the interference between different measurement standards between control variables can be effectively eliminated, thereby improving the accuracy of the judgment result of the operating stage to which the first operating data belongs. Moreover, the calculation logic of the whole calculation process is simple and can effectively improve the calculation speed.

[0126] In some embodiments, the membership degree between the first running data and each running stage is determined based on the first running data and multiple Mahalanobis distances. Specifically, the membership degree of the first running data corresponding to each running stage can be determined based on multiple Mahalanobis distances and membership degree calculation formulas. The membership degree calculation formula can be as shown in formula (2).

[0127]

[0128] in, The Mahalanobis distance for each running stage corresponds to the first running data. For the first running data corresponding to the first i Membership degree of each operational phase The Mahalanobis distance for each running stage corresponds to the first running data. M The total number of operation phases. M positive integer, m As a preset constant, m >1.

[0129] As a concrete example, ∈[0,1], and .

[0130] In this embodiment of the application, by combining the second operating data that divides the operation of different wind turbine generators, and based on the idea of ​​fuzzy clustering, the Mahalanobis distance between the first operating data and each operating stage can be obtained. This can effectively eliminate the interference of different measurement standards between control variables, and then use the Mahalanobis distance to determine the membership degree of the first operating data to different stages, thereby improving the accuracy of the judgment result of the operating stage to which the first operating data belongs.

[0131] Therefore, after calculating the membership degree through step 120, step 130 can be executed.

[0132] In step 130, the running stage corresponding to the membership degree that meets the preset conditions is determined as the target running stage corresponding to the first running data.

[0133] In this embodiment of the application, since the wind turbine generator set includes multiple operating stages, when multiple second operating data are obtained for each of the multiple operating stages, a membership degree can be obtained for each operating stage.

[0134] Therefore, screening criteria can be preset. For example, the preset criteria could be determining the largest membership degree among all membership degrees. Correspondingly, when determining the target operating stage, the operating stage corresponding to the largest membership degree is identified as the target operating stage corresponding to the first operating data. In this way, by determining the operating stage to which the first operating data belongs, the fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, which helps improve the detection accuracy of faults in wind turbine generators.

[0135] After obtaining the target running stage corresponding to the first running data according to step 130, step 140 can be executed next.

[0136] In step 140, the fault detection result of the first operating data is determined according to the preset fault analysis model corresponding to the target operating stage.

[0137] In some embodiments, the preset fault analysis model may include, for example, a preset machine learning algorithm, a threshold analysis model, etc. For example, the preset machine learning algorithm may be trained using training samples corresponding to the target operating stage; the threshold analysis model may be, for example, a fault alarm threshold set according to the preset algorithm corresponding to the target operating stage, etc., without specific limitations.

[0138] According to the embodiments of this application, by using a fault analysis model corresponding to the target operating stage to perform corresponding fault detection, the detection accuracy of fault detection of wind turbine generator sets can be effectively improved.

[0139] In some embodiments, the operation of a wind turbine generator may consist of two adjacent operation phases, combined with Figure 4 As shown, adjacent operating phases include AB, BC, and CD. That is, the first operating data may be data collected during different periods of change in the operating phases. To better analyze the first operating data, optionally, determining the target operating phase corresponding to the first operating data may include the following steps:

[0140] First, calculate the harmonic mean of the membership degrees of two adjacent running stages in multiple running stages; then, if the harmonic mean meets the preset threshold range, determine that the target running stage includes two adjacent running stages.

[0141] Optionally, the harmonic mean of the membership degrees of two adjacent operating stages in multiple operating stages can be calculated according to formula (3).

[0142] (3)

[0143] in, This is the harmonic mean of the membership degrees corresponding to two adjacent operational phases. For the first running data corresponding to the first i Membership degree of each operational phase.

[0144] As a concrete example, when the harmonic mean meets a preset threshold range, the first operating data can be determined to be data collected during the period of change in adjacent operating stages. Optionally, the preset threshold range is [0.4, 0.5]. Thus, it can be determined that the target operating stage includes two adjacent operating stages, which helps improve the accuracy of subsequent fault detection of the wind turbine generator.

[0145] In some embodiments, when the target operating phase includes two adjacent operating phases, step 140 of this application embodiment can specifically determine the fault detection result of the first operating data based on a preset fault analysis model corresponding to any one of the two adjacent operating phases.

[0146] Since the first operational data is collected during the transition between adjacent operational phases, it can include data features corresponding to the two operational phases. Therefore, a preset fault analysis model corresponding to either of the two adjacent operational phases can be selected. For example, a preset fault analysis model with higher fault detection accuracy can be selected based on the fault detection accuracy requirements; alternatively, a preset fault analysis model that outputs fault detection results faster can be selected based on the fault detection speed requirements. No specific limitations are imposed here.

[0147] According to the embodiments of this application, by dividing the operating data into stages, a fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, thereby effectively improving the detection accuracy of fault detection in wind turbine generator sets.

[0148] As a specific example, let's take the detection of whether there is an abnormality in the pitch angle of the pitch system in a wind turbine generator set.

[0149] If the operating stage corresponding to the first pair of operating data is determined to be stage A or stage B, that is, the first operating data was collected when the wind turbine was in the startup stage or the optimal wind energy tracking stage, then the pitch angle of the pitch system can be determined to be abnormal by performing probability fitting based on kernel density estimation and Gaussian mixture model, and by setting a preset detection threshold according to the distribution function. For example, if the operating stage of the first pair of operating data is determined to be stage C or stage D, that is, the first operating data was collected when the wind turbine was in the rated speed stage or the rated power stage, then, since the pitch angle has a non-linear relationship with wind speed, the pitch angle of the pitch system can be determined to be abnormal by performing boundary fitting based on single-class support vector machine, and by setting a preset detection threshold corresponding to the boundary fitting function.

[0150] For example, when the harmonic mean The first set of operational data was collected during the transition from the optimal wind energy tracking phase to the rated speed phase of the wind turbine generator. Accordingly, the first set of operational data only needs to be used to determine whether the pitch angle of the pitch system is abnormal, based on either the preset detection threshold corresponding to the probability fitting function or the preset detection threshold corresponding to the boundary fitting function.

[0151] In another example, the pitch angle can be fitted using the eXtreme Gradient Boosting (XGBoost) algorithm, with different upper limits for residuals set for different stages, to determine whether there are any anomalies in the pitch angle of the pitch system.

[0152] It is understood that the above method for detecting whether there is an abnormality in the pitch angle of the pitch system in a wind turbine generator is merely an example and does not constitute a specific limitation on this application.

[0153] Figure 5 This is a schematic diagram of the structure of a wind turbine generator fault detection device provided in an embodiment of this application, as shown below. Figure 5 As shown, the wind turbine generator fault detection device 500 may include: an acquisition module 510 and a processing module 520.

[0154] The acquisition module 510 is used to acquire the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period. The wind turbine generator set includes multiple operating stages, and the operating parameter dataset includes multiple second operating data corresponding to each operating stage.

[0155] The processing module 520 is used to determine the membership degree of the first running data to each running stage based on the first running data and multiple second running data corresponding to each running stage;

[0156] The processing module 520 is also used to determine the running stage corresponding to the membership degree that meets the preset conditions, which is the target running stage corresponding to the first running data;

[0157] The processing module 520 is also used to determine the fault detection result of the first operating data according to the preset fault analysis model corresponding to the target operating stage.

[0158] The wind turbine generator set fault detection device of this application acquires first operating data and operating parameter dataset of the wind turbine generator set within a preset time period. Since the wind turbine generator set includes multiple operating stages, the operating parameter dataset includes multiple second operating data corresponding to each operating stage. Next, based on the first operating data and the multiple second operating data corresponding to each operating stage, the membership degree of the first operating data to each operating stage can be determined; and the operating stage corresponding to the membership degree that meets the preset conditions can be determined as the target operating stage corresponding to the first operating data. In this way, the control logic experience of the wind turbine generator set in different operating stages can be combined to determine the target operating stage corresponding to the first operating data. Next, based on the preset fault analysis model corresponding to the target operating stage, the fault detection result of the first operating data is determined. Since the operating data is divided into stages, the fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, thereby effectively improving the detection accuracy of the wind turbine generator set fault detection.

[0159] In some embodiments, the processing module 520 is further configured to determine the Mahalanobis distance of each running stage corresponding to the first running data based on the first running data and a plurality of second running data corresponding to each running stage;

[0160] The processing module 520 is also used to determine the membership degree between the first running data and each running stage based on the first running data and multiple Mahalanobis distances.

[0161] In this way, by calculating the Mahalanobis distance of the first running data for each running stage, the interference of different measurement standards between variables can be effectively eliminated, thereby improving the accuracy of the judgment result of the running stage to which the first running data belongs. Moreover, the calculation logic of the whole calculation process is simple and can effectively improve the calculation speed.

[0162] In some embodiments, the processing module 520 is further configured to determine the covariance matrix and mean corresponding to each running stage based on the multiple second running data corresponding to each running stage;

[0163] The processing module 520 is also used to determine the Mahalanobis distance for each running stage corresponding to the first running data based on the first running data and the covariance matrix and mean corresponding to each running stage.

[0164] Thus, by calculating the Mahalanobis distance for each running stage corresponding to the first running data, the interference of different measurement standards between variables can be effectively eliminated, which is conducive to improving the accuracy of the judgment result of the running stage to which the first running data belongs.

[0165] In some embodiments, the processing module 520 is further configured to determine the membership degree of the first running data corresponding to each running stage based on multiple Mahalanobis distance and membership degree calculation formulas, wherein the membership degree calculation formula is:

[0166]

[0167] in, The Mahalanobis distance for each running stage corresponds to the first running data. For the first running data corresponding to the first i Membership degree of each operational phase The Mahalanobis distance for each running stage corresponds to the first running data. M The total number of operation phases. M positive integer, m As a preset constant, m >1.

[0168] In this embodiment of the application, by combining the second operating data that divides the operation of different wind turbine generators, and based on the idea of ​​fuzzy clustering, the Mahalanobis distance between the first operating data and each operating stage can be obtained. This can effectively eliminate the interference of different measurement standards between variables, and then use the Mahalanobis distance to determine the membership degree of the first operating data to different stages, thereby improving the accuracy of the judgment result of the operating stage to which the first operating data belongs.

[0169] In some embodiments, the processing module 520 is further configured to determine the running stage corresponding to the largest membership degree among multiple membership degrees, which is the target running stage corresponding to the first running data.

[0170] According to the embodiments of this application, by using a fault analysis model corresponding to the target operating stage to perform corresponding fault detection, the detection accuracy of fault detection of wind turbine generator sets can be effectively improved.

[0171] In some embodiments, the plurality of operation phases includes two adjacent operation phases;

[0172] The processing module 520 is also used to calculate the harmonic mean of the membership degrees of two adjacent running stages in multiple running stages;

[0173] The processing module 520 is also used to determine that the target running stage includes two adjacent running stages, provided that the harmonic average value meets the preset threshold range.

[0174] In this way, it can be determined that the target operating phase includes two adjacent operating phases, which is beneficial to improving the accuracy of subsequent fault detection of wind turbine generators.

[0175] In some embodiments, the processing module 520 is further configured to determine the fault detection result of the first operating data based on a preset fault analysis model corresponding to any one of two adjacent operating phases.

[0176] According to the embodiments of this application, by dividing the operating data into stages, a fault analysis model corresponding to the target operating stage can be selected for targeted data analysis, thereby effectively improving the detection accuracy of fault detection in wind turbine generator sets.

[0177] In some embodiments, the acquisition module 510 is further configured to acquire multiple third operating data of the wind turbine generator set within a preset time period, wherein each third operating data includes the rotational speed of the wind turbine generator set, the torque of the wind turbine generator set, and the pitch angle of the wind turbine generator set.

[0178] The processing module 520 is also used to perform preset statistical analysis on multiple speeds, multiple torques, and multiple pitch angles of the wind turbine generator set, determine the operating stage corresponding to each third operating data, and obtain multiple second operating data corresponding to each operating stage.

[0179] In this way, the operating status information of the wind turbine generator corresponding to the operating data generated by the wind turbine generator within a preset time period can be accurately determined, so as to make full use of the data characteristics of different operating stages.

[0180] In some embodiments, the processing module 520 is further configured to filter out third operating data that do not meet preset wind turbine operating conditions from a plurality of third operating data of the wind turbine generator set.

[0181] According to the embodiments of this application, based on preset wind turbine generator operating conditions, the corresponding operating data under the healthy and stable state of the wind turbine generator can be screened to improve the reliability of the extracted operating parameter dataset.

[0182] In some embodiments, the multiple operation phases include a startup phase and an optimal wind energy tracking phase;

[0183] The processing module 520 is also used to divide multiple speeds of the wind turbine generator set according to the preset speed division rules to obtain multiple speed compartments;

[0184] The processing module 520 is also used to determine the mean torque in each speed chamber, the dispersion of torque in each speed chamber, and to determine the first fitting curve of the mean torque and the corresponding speed of each speed chamber, and the second fitting curve of the dispersion of torque and the corresponding speed of each speed chamber.

[0185] The processing module 520 is also used to determine whether the operating stage corresponding to each third operating data is the start-up stage or the optimal wind energy tracking stage based on the tangent change rate of the first fitted curve and the tangent change rate of the second fitted curve.

[0186] Thus, according to the operation stage division method provided in the embodiments of this application, by analyzing the statistical indicators of torque corresponding to different speed chambers, the data characteristics of different operation stages can be understood, thereby accurately determining the operation stage corresponding to the third operation data.

[0187] In some embodiments, the multiple operating phases also include a rated speed phase and a rated power phase, and the third operating data also includes multiple operating powers of the wind turbine generator set;

[0188] The processing module 520 is also used to divide the multiple operating powers of the wind turbine generator set according to the preset power division rules to obtain multiple power compartments;

[0189] The processing module 520 is also used to determine the maximum value of the pitch angle in each power compartment, and to determine the third fitting curve of the maximum value of the pitch angle and the corresponding power of each power compartment.

[0190] The processing module 520 is also used to determine whether the operating stage corresponding to each third operating data is the rated speed stage or the rated power stage based on the tangent change rate of the third fitted curve.

[0191] Therefore, based on the above analysis of the third operational data, the operational stages corresponding to the third operational data can be further refined, thereby improving the accuracy of judging the operational stages corresponding to the third operational data.

[0192] It is understood that the wind turbine generator set fault detection device 500 in this application embodiment can correspond to the execution subject of the wind turbine generator set fault detection method provided in this application embodiment. The specific details of the operation and / or function of each module / unit of the wind turbine generator set fault detection device 500 can be found in the description of the corresponding part in the wind turbine generator set fault detection method provided in the above application embodiment. For the sake of brevity, it will not be repeated here.

[0193] Figure 6 A schematic diagram of a wind turbine generator fault detection device according to an embodiment of this application is shown. Figure 6As shown, the device may include a processor 601 and a memory 602 storing computer program instructions.

[0194] Specifically, the processor 601 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0195] Memory 602 may include a mass storage device for information or instructions. For example, and not limitingly, memory 602 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 602 may include removable or non-removable (or fixed) media, or memory 602 may be a non-volatile solid-state memory. Memory 602 may be internal or external to equipment for wind turbine fault detection.

[0196] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0197] The processor 601 reads and executes computer program instructions stored in the memory 602 to implement the method described in the embodiments of this application and achieve the corresponding technical effects achieved by executing the method in the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0198] In one example, the wind turbine fault detection device may further include a communication interface 603 and a bus 610. For example, Figure 6 As shown, the processor 601, memory 602, and communication interface 603 are connected through bus 610 and complete communication with each other.

[0199] The communication interface 603 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0200] Bus 610 includes hardware, software, or both, that couples components of an online information flow metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 610 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0201] The equipment for detecting wind turbine generator faults can execute the method for detecting wind turbine generator faults in the embodiments of this application, thereby achieving the corresponding technical effects of the method for detecting wind turbine generator faults described in the embodiments of this application.

[0202] Furthermore, in conjunction with the wind turbine generator fault detection method in the above embodiments, this application embodiment can provide a readable storage medium for implementation. This readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the wind turbine generator fault detection methods in the above embodiments. Examples of readable storage media can be non-transitory machine-readable media, such as electronic circuits, semiconductor memory devices, read-only memory (ROM), floppy disks, compact disc read-only memory (CD-ROM), optical discs, hard disks, etc.

[0203] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0204] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact disc read-only memory (CD-ROM), optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0205] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0206] Furthermore, in conjunction with the wind turbine generator set fault detection method and apparatus described in the above embodiments, as well as the readable storage medium, this application embodiment can provide a computer program product for implementation. When the instructions in the computer program product are executed by the processor of an electronic device, the electronic device causes the electronic device to perform any of the wind turbine generator set fault detection methods described in the above embodiments.

[0207] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, 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, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0208] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for fault detection in wind turbine generator sets, characterized in that, include: Acquire first operating data of the wind turbine generator set and an operating parameter dataset of the wind turbine generator set within a preset time period, wherein the wind turbine generator set includes multiple operating stages, and the operating parameter dataset includes multiple second operating data corresponding to each operating stage; the multiple operating stages include two adjacent operating stages; Based on the first running data and multiple second running data corresponding to each running stage, determine the membership degree of the first running data corresponding to each running stage; The operational stage corresponding to the membership degree that meets the preset conditions is determined as the target operational stage corresponding to the first operational data; Based on the preset fault analysis model corresponding to the target operation stage, the fault detection result of the first operation data is determined; The step of determining the operational stage corresponding to the membership degree that meets the preset conditions as the target operational stage corresponding to the first operational data includes: Calculate the harmonic mean of the membership degrees of two adjacent operating stages among the plurality of operating stages; If the harmonic average value meets the preset threshold range, the target operating phase is determined to include the two adjacent operating phases.

2. The method according to claim 1, characterized in that, The step of determining the membership degree of the first running data to each running stage based on the first running data and multiple second running data corresponding to each running stage includes: Based on the first running data and multiple second running data corresponding to each running stage, determine the Mahalanobis distance of the first running data for each running stage; Based on the first running data and multiple Mahalanobis distances, the membership degree of the first running data to each running stage is determined.

3. The method according to claim 2, characterized in that, The step of determining the Mahalanobis distance for each running stage based on the first running data and multiple second running data corresponding to each running stage includes: Based on multiple second-stage data corresponding to each stage of operation, determine the covariance matrix and mean for each stage of operation. Based on the first running data and the covariance matrix and mean corresponding to each running stage, the Mahalanobis distance corresponding to the first running data for each running stage is determined.

4. The method according to claim 2 or 3, characterized in that, The step of determining the membership degree between the first running data and each running stage based on the first running data and multiple Mahalanobis distances includes: Based on the aforementioned Mahalanobis distance and membership degree calculation formulas, the membership degree of the first running data corresponding to each running stage is determined, wherein the membership degree calculation formula is: in, The Mahalanobis distance for each of the aforementioned running stages corresponding to the first running data. For the first running data corresponding to the first i Membership degree of each of the aforementioned operational phases The Mahalanobis distance for each of the aforementioned running stages corresponding to the first running data. M The total number of the aforementioned operation phases. M positive integer, m As a preset constant, m >1.

5. The method according to claim 1, characterized in that, The step of determining the operational stage corresponding to the membership degree that meets the preset conditions as the target operational stage corresponding to the first operational data includes: The operational stage corresponding to the largest membership degree among the multiple membership degrees is determined as the target operational stage corresponding to the first operational data.

6. The method according to claim 1, characterized in that, The step of determining the fault detection result of the first operating data according to the preset fault analysis model corresponding to the target operating stage includes: Based on the preset fault analysis model corresponding to any one of the two adjacent operating phases, the fault detection result of the first operating data is determined.

7. The method according to claim 1, characterized in that, Before acquiring the first operating data of the wind turbine generator set and the operating parameter dataset of the wind turbine generator set within a preset time period, the method further includes: Acquire multiple third operating data of the wind turbine generator set within the preset time period, wherein each third operating data includes the rotational speed, torque, and pitch angle of the wind turbine generator set; Pre-set statistical analysis is performed on multiple rotational speeds, multiple torques, and multiple pitch angles of the wind turbine generator set to determine the operating stage corresponding to each of the third operating data, and multiple second operating data corresponding to each operating stage are obtained.

8. The method according to claim 7, characterized in that, The acquisition of multiple third-party operating data of the wind turbine generator set within the preset time period includes: Third operating data that does not meet the preset operating conditions of the wind turbine generator set are filtered out from multiple third operating data of the wind turbine generator set.

9. The method according to claim 7, characterized in that, The multiple operating phases include a startup phase and an optimal wind energy tracking phase; the pre-defined statistical analysis of multiple rotational speeds, multiple torques, and multiple pitch angles of the wind turbine generator set to determine the operating phase corresponding to each of the third operating data includes: According to the preset speed division rules, the multiple speeds of the wind turbine generator set are divided to obtain multiple speed compartments; The mean torque in each speed chamber, the dispersion of torque in each speed chamber, and the first fitting curve of the mean torque with the corresponding speed of each speed chamber, and the second fitting curve of the dispersion of torque with the corresponding speed of each speed chamber are determined. Based on the tangent change rate of the first fitted curve and the tangent change rate of the second fitted curve, the operating stage corresponding to each of the third operating data is determined to be either the start-up stage or the optimal wind energy tracking stage.

10. The method according to claim 9, characterized in that, The multiple operating stages also include a rated speed stage and a rated power stage, and the third operating data also includes multiple operating power levels of the wind turbine generator set; the step of performing a preset statistical analysis on multiple speeds, multiple torques, and multiple pitch angles of the wind turbine generator set to determine the operating stage corresponding to each of the third operating data points includes: According to the preset power division rules, the multiple operating powers of the wind turbine generator set are divided to obtain multiple power compartments; Determine the maximum value of the pitch angle in each power compartment, and determine the third fitting curve of the maximum value of the pitch angle and the corresponding power of each power compartment; Based on the rate of change of the tangent of the third fitted curve, the operating stage corresponding to each of the third operating data is determined to be either the rated speed stage or the rated power stage.

11. A device for fault detection of wind turbine generator sets, characterized in that, The device includes: The acquisition module is used to acquire first operating data of the wind turbine generator set and an operating parameter dataset of the wind turbine generator set within a preset time period. The wind turbine generator set includes multiple operating stages, and the operating parameter dataset includes multiple second operating data corresponding to each operating stage. The multiple operating stages include two adjacent operating stages. The processing module is configured to determine the membership degree of the first running data corresponding to each running stage based on the first running data and multiple second running data corresponding to each running stage; The processing module is further configured to determine the running stage corresponding to the membership degree that meets the preset conditions as the target running stage corresponding to the first running data; The processing module is further configured to determine the fault detection result of the first operating data according to a preset fault analysis model corresponding to the target operating stage; The processing module is further configured to calculate the harmonic mean of the membership degrees of two adjacent running stages among the plurality of running stages; and, if the harmonic mean satisfies a preset threshold range, determine that the target running stage includes the two adjacent running stages.

12. A device for fault detection of wind turbine generator sets, characterized in that, The device includes: a processor and a memory storing computer program instructions; The processor reads and executes the computer program instructions to implement the method for wind turbine generator set fault detection as described in any one of claims 1-10.

13. A readable storage medium, characterized in that, The readable storage medium stores computer program instructions, which, when executed by a processor, implement the method for wind turbine generator fault detection as described in any one of claims 1-10.

14. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device performs the method for wind turbine generator fault detection as described in any one of claims 1-10.

Images