Variable pitch system fault positioning method based on wind turbine SCADA data and related device
By combining data cleaning and local anomaly factor algorithms with deep learning models, the problems of noise and outliers in SCADA data were solved, accurate positioning and early warning of pitch system failures were achieved, and the operating stability and maintenance efficiency of wind turbines were improved.
Patent Information
- Application Number
- CN202510825662.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-19
- Publication Date
- 2025-09-23
AI Technical Summary
In the existing technology, the variable pitch system fault location based on SCADA data has problems such as noise interference, data missing and outliers, resulting in low positioning accuracy, difficulty in adapting to complex and changeable fault modes, and weak algorithm generalization ability, which cannot meet actual needs.
A method combining data cleaning and local anomaly factor algorithm is adopted to improve data quality through standardization and missing value processing. The support vector single classification data cleaning model is used to identify abnormal data. Combined with the deep learning fault diagnosis model, a fault tree model is constructed for accurate fault location.
It significantly improves data quality and fault location accuracy, achieves early warning and rapid fault location, reduces operation and maintenance costs, and improves the operating stability and maintenance efficiency of wind turbines.
Smart Images
Figure CN120684369A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault location, and in particular to a method for locating a fault of a variable pitch system based on wind turbine SCADA data and a related device. Background Art
[0002] As an important executive device for controlling and protecting wind turbines, the variable pitch system plays a vital role in the safe, stable and efficient operation of the unit. It achieves efficient capture of wind energy and unit protection by dynamically adjusting the pitch angle. Its stable operation is directly related to power generation efficiency and equipment safety. Once the variable pitch system fails, it will not only reduce power generation, but may also trigger a chain reaction, threatening the overall safety of the unit. Therefore, fast and accurate fault location is the key to ensuring the reliable operation of wind turbines.
[0003] However, fault location based on SCADA data currently faces numerous challenges. The massive amounts of data collected by SCADA systems are often plagued by noise, missing data, and outliers. These data quality defects can interfere with the accuracy of fault location algorithms, leading to incorrect fault identification and location, resulting in low location accuracy. Furthermore, traditional fault location methods struggle to capture complex and changing fault patterns, and their algorithms lack generalization capabilities and adaptability to diverse operating conditions and environments, making fault location accuracy insufficient to meet practical requirements. Summary of the Invention The purpose of the present invention is to provide a variable pitch system fault location method and related devices based on wind turbine SCADA data to overcome the problems existing in the prior art. The present invention can effectively remove noise, missing values and outliers in the SCADA system operation data, significantly improve data quality, and automatically extract complex and hidden fault features from the feature vectors of abnormal operation data, accurately identify various types of variable pitch system faults, and quickly locate the fault location.
[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for locating a fault in a pitch system of a wind turbine based on SCADA data, comprising the following steps: Step 1: Collect the variable pitch system operation data of the wind turbine SCADA system and clean the operation data; Step 2: Use the local abnormality factor algorithm to monitor the operating data after cleaning in real time to determine whether there is an abnormality in the pitch system. If there is an abnormality in the pitch system, execute step 3; if there is no abnormality in the pitch system, execute step 1 again. Step 3: Extract the feature vector of the abnormal operation data and input the feature vector into the fault diagnosis model based on deep learning to diagnose and locate the fault; Furthermore, the operation data is cleaned in step 1, specifically including: The running data is standardized and missing value processed in sequence to obtain normal data and abnormal data. A support vector single classification data cleaning model is constructed based on the normal data. The support vector single classification data cleaning model is trained based on the normal data and abnormal data. The trained support vector single classification data cleaning model is used to clean the running data. Furthermore, the standardization process specifically includes: scaling the operating data to a uniform scale range; The missing value processing specifically includes: using a time series-based interpolation algorithm to fill the missing values of the normalized operating data, marking the operating data whose missing values do not exceed a preset threshold as normal data, and marking the operating data whose missing values exceed a preset threshold as abnormal data; Furthermore, the support vector single classification data cleaning model is trained based on normal data and abnormal data, and the running data is cleaned using the trained support vector single classification data cleaning model, specifically including: Normal data and abnormal data are used as prepared data, the prepared data are divided into a training set and a test set, various parameters of the support vector single classification data cleaning model are set, the support vector single classification data cleaning model is trained using the training set, and then various indicators of the support vector single classification data cleaning model are calculated using the test set. The trained support vector single classification data cleaning model is evaluated according to the various indicators, and then the parameters and data volume of the support vector single classification data cleaning model are tuned according to the evaluation results to obtain a trained support vector single classification data cleaning model; Input the running data into the trained support vector single classification data cleaning model, and use the support vector single classification data cleaning model to identify the running data. If it is identified as normal data, the normal data is directly output; if it is identified as abnormal data, the abnormal data is cleaned, wherein the cleaning includes: using a smoothing correction algorithm based on adjacent data for cleaning, or deleting the abnormal data; Furthermore, in step 2, a local anomaly factor algorithm is used to perform real-time monitoring on the cleaned operating data, specifically including: The local abnormality factor algorithm is used to calculate the local abnormality factor score in the cleaned operating data. The abnormality threshold is set according to the local abnormality factor score. When the local abnormality factor score is greater than or equal to the abnormality threshold, it is determined that the pitch system has an abnormality; when the local abnormality factor score is less than the abnormality threshold, it is determined that the pitch system does not have an abnormality. Furthermore, the feature vector of the abnormal operation data in step 3 includes: a time domain feature vector, a frequency domain feature vector and a time-frequency domain feature vector; Furthermore, the fault diagnosis and location in step three specifically include: If a fault is diagnosed in the pitch system, a fault tree model is established based on the fault type, combined with the structure and working principle of the pitch system. The fault tree model uses the fault type as the top event, analyzes the intermediate events and bottom events that lead to the fault, and determines the specific location of the fault by calculating the probability and importance of the bottom event. If it is diagnosed that there is no fault in the pitch system, perform step 1 again.
[0005] In a second aspect, the present invention provides a pitch system fault location system based on wind turbine SCADA data, comprising: Data acquisition and cleaning module, used to collect the variable pitch system operating data of the wind turbine SCADA system and clean the operating data; The data monitoring module is used to monitor the operating data after cleaning in real time using a local abnormality factor algorithm to determine whether there is any abnormality in the pitch system; The fault diagnosis and location module is used to extract the feature vectors of abnormal operation data, input the feature vectors into the fault diagnosis model based on deep learning, and diagnose and locate the fault.
[0006] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.
[0007] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.
[0008] The above technical solution has the following advantages or beneficial effects: First, the present invention provides a method for locating a pitch system fault based on wind turbine SCADA data. By combining data cleaning with a local anomaly factor algorithm, a solid foundation is laid for fault location. Data cleaning can effectively remove noise, missing values, and outliers in the SCADA system operation data, significantly improving data quality and the accuracy of fault location. The local anomaly factor algorithm can monitor data in real time and quickly identify pitch system anomalies. By calculating the distance between data points and neighboring points to measure the degree of anomaly, early warning of faults is achieved, thereby avoiding fault escalation and reducing power generation losses caused by fault downtime. The application of a deep learning fault diagnosis model greatly enhances the accuracy of fault diagnosis and location. The deep learning model has a strong feature learning ability and can automatically extract complex and hidden fault features from the feature vectors of abnormal operation data, accurately identify various types of pitch system faults, and quickly locate the fault location. In addition, the present method realizes the automation of the entire process from data collection, anomaly monitoring to fault diagnosis and location, greatly improving fault location efficiency, reducing the workload and labor costs of operation and maintenance personnel, providing a strong guarantee for the stable operation and efficient maintenance of wind turbines, and is of great significance to promoting the sustainable development of the wind power industry.
[0009] Furthermore, the pre-steps of standardization and missing value processing effectively standardize the data format and integrity. Standardization can eliminate dimensional differences between data, making data of different dimensions comparable and laying a unified foundation for subsequent analysis. Missing value processing avoids analytical bias caused by incomplete data and reduces information loss. On this basis, a support vector single-classification data cleaning model is constructed and trained to accurately identify abnormal data. Through training with a large amount of normal data and abnormal data, the model can adapt to complex data patterns, distinguish between noise and valid data with high accuracy, and effectively filter out interference information. The trained model cleans the operating data, which can significantly improve data quality, reduce the interference of data noise on the fault location algorithm, and provide high-quality data support for subsequent abnormal monitoring and fault diagnosis. Ultimately, it ensures that the fault location results are more accurate, reduces misjudgments and missed judgments, improves the reliability and efficiency of wind turbine pitch system fault diagnosis, reduces operation and maintenance costs, and ensures stable operation of the unit.
[0010] Furthermore, the data cleaning solution significantly improves the accuracy and reliability of data processing through scientific model training and optimization strategies. It divides the data into training sets and test sets, and dynamically tunes the model parameters and data volume in combination with indicator evaluation, so that the support vector single classification model can accurately learn the feature boundaries of normal data and effectively identify abnormal data. The dual-mode cleaning of abnormal data based on the smoothing correction algorithm and deletion strategy of adjacent data not only ensures the continuity and availability of data, but also avoids noise interference. It can significantly improve the quality of the cleaned SCADA data, provide a more accurate data source for subsequent fault location, reduce the risk of misjudgment, and improve the efficiency of fault diagnosis. It is of great significance to ensure the stable operation of wind turbines and reduce operation and maintenance costs.
[0011] Furthermore, the local anomaly factor algorithm can accurately calculate the degree of data anomaly based on the density difference between the data point and its neighboring points, and dynamically set the anomaly threshold so that it can adapt to different operating conditions; it can quickly identify various types of sudden anomalies without pre-setting the fault mode, and has greater versatility and flexibility than traditional methods; by calculating the local anomaly factor score in real time, it can issue timely warnings at the incipient stage of the fault, avoid the expansion of the fault, effectively reduce downtime and maintenance costs, significantly improve the safety and stability of wind turbine operation, and provide strong guarantees for the efficient operation and maintenance of wind power systems.
[0012] Furthermore, by constructing a fault tree model with the fault type as the top event, the cause-effect relationship of the variable pitch system failure can be systematically sorted out, and the fault transmission path can be analyzed from the structural and working principle levels; by calculating the probability and importance of the bottom event, the contribution of each potential fault source to the top event can be quantified, thereby accurately locking the specific location of the fault. Compared with a single diagnostic method, this strategy not only improves the accuracy of fault location, but also enhances the logic and traceability of fault analysis through a visual fault tree structure, providing a scientific basis for operation and maintenance personnel to formulate targeted maintenance plans, effectively shortening fault handling time and reducing operation and maintenance costs.
[0013] Secondly, the present invention provides a variable pitch system fault location system based on wind turbine SCADA data. The data acquisition and cleaning module ensures the accuracy and integrity of the input data. By collecting and cleaning the variable pitch system operation data of the SCADA system, the noise and redundant information are effectively removed, providing a high-quality data basis for subsequent analysis, and improving the reliability of fault location; the data monitoring module adopts a local anomaly factor algorithm for real-time monitoring, which can quickly and accurately determine whether there is an abnormality in the variable pitch system, realize early detection of faults, avoid further deterioration of faults, and reduce maintenance costs and downtime; the fault diagnosis and location module uses a deep learning model to analyze the abnormal data feature vector, which can accurately diagnose the fault type and locate the fault location, provide clear guidance for maintenance personnel, and greatly improve maintenance efficiency; overall, the system forms a complete closed loop from data acquisition, monitoring to fault diagnosis and location, which significantly improves the operating stability and maintenance efficiency of the wind turbine variable pitch system.
[0014] In a third aspect, the present invention provides a computer device that can efficiently implement the steps of the method of the present invention by executing a specific computer program through a processor. When performing data processing tasks, the computer device can accurately perform numerical calculations and logical judgments, avoiding errors caused by human factors; at the same time, since the computer program has a high degree of stability and reliability, the accuracy and consistency of the data processing results can be ensured.
[0015] In a fourth aspect, the present invention provides a computer-readable storage medium. By programming the steps of the method of the present invention into a computer program and storing it on a computer-readable storage medium, users can easily load these programs onto any compatible computer device and execute them without rewriting or converting the code, thereby greatly improving the convenience and flexibility of program execution. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 Schematic diagram of a flow chart of a method for locating a fault in a variable pitch system of a wind turbine based on SCADA data of the present invention; Figure 2 Schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION
[0017] The present invention will be further described in detail below with reference to specific embodiments, which are intended to explain the present invention rather than to limit it. In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention. It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0018] Example: See also Figure 1 The present invention provides a method for locating faults in a variable pitch system of a wind turbine based on SCADA data, comprising the following steps: Step 1: Collect the variable pitch system operation data of the wind turbine SCADA system and clean the operation data; Specifically, the operation data is cleaned, including: performing standardization processing and missing value processing on the operation data in sequence, wherein the standardization processing includes: scaling the operation data to a uniform scale range, and the missing value processing includes: filling the missing values of the normalized operation data using a time series-based interpolation algorithm, marking the operation data whose missing values do not exceed a preset threshold as normal data, and marking the operation data whose missing values exceed the preset threshold as abnormal data; obtaining normal data and abnormal data, constructing a support vector single-classification data cleaning model based on the normal data, training the support vector single-classification data cleaning model based on the normal data and the abnormal data, and using the trained support vector single-classification data cleaning model to clean the operation data; Specifically, a support vector single classification data cleaning model is trained based on normal data and abnormal data, and the trained support vector single classification data cleaning model is used to clean the running data, including: taking normal data and abnormal data as prepared data, dividing the prepared data into a training set and a test set, setting various parameters of the support vector single classification data cleaning model, using the training set to train the support vector single classification data cleaning model, and then using the test set to calculate various indicators of the support vector single classification data cleaning model, evaluating the trained support vector single classification data cleaning model according to the various indicators, and then tuning the parameters and data volume of the support vector single classification data cleaning model according to the evaluation results to obtain a trained support vector single classification data cleaning model; inputting the running data into the trained support vector single classification data cleaning model, identifying the running data through the support vector single classification data cleaning model, and directly outputting the normal data if it is identified as normal data; and cleaning the abnormal data if it is identified as abnormal data, wherein the cleaning includes: using a smoothing correction algorithm based on adjacent data for cleaning, or deleting the abnormal data; Step 2: Use the local anomaly factor algorithm to monitor the cleaned operating data in real time, use the local anomaly factor algorithm to calculate the local anomaly factor score in the cleaned operating data, set an anomaly threshold based on the local anomaly factor score, and when the local anomaly factor score is greater than or equal to the anomaly threshold, determine that the pitch system is abnormal, and execute step 3; When the local abnormality factor score is less than the abnormality threshold, it is determined that there is no abnormality in the pitch system, and step 1 is executed again; Step three: Extract the feature vector of the abnormal operation data and input the feature vector into the fault diagnosis model based on deep learning to diagnose and locate the fault. If a fault is diagnosed in the variable pitch system, a fault tree model is established based on the fault type, combined with the structure and working principle of the variable pitch system. The fault tree model uses the fault type as the top event, analyzes the intermediate events and bottom events that lead to the fault, and determines the specific location of the fault by calculating the probability and importance of the bottom event. If it is diagnosed that there is no fault in the pitch system, perform step 1 again; Specifically, the feature vectors of abnormal operation data include: time domain feature vectors, frequency domain feature vectors, and time-frequency domain feature vectors.
[0019] In one embodiment of the present invention, a pitch system fault location system based on wind turbine SCADA data is provided, comprising: Data acquisition and cleaning module, used to collect the variable pitch system operating data of the wind turbine SCADA system and clean the operating data; The data monitoring module is used to monitor the operating data after cleaning in real time using a local abnormality factor algorithm to determine whether there is any abnormality in the pitch system; The fault diagnosis and location module is used to extract the feature vectors of abnormal operation data, input the feature vectors into the fault diagnosis model based on deep learning, and diagnose and locate the fault.
[0020] See also Figure 2 In one embodiment of the present invention, a computer device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium; the processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which is the computing core and control core of the terminal, and is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method process or corresponding function; the processor described in the embodiment of the present invention can be used to operate the variable pitch system fault location method based on wind turbine SCADA data.
[0021] In one embodiment of the present invention, a computer-readable storage medium is provided, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium herein may include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides storage space, which stores the operating system of the terminal. In addition, the storage space also stores one or more instructions suitable for being loaded and executed by the processor. These instructions may be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium herein may be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor may load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the pitch system fault location method based on wind turbine SCADA data in the embodiment.
[0022] Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0023] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts 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, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate the instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0024] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0025] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0026] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for locating a fault in a variable pitch system of a wind turbine based on SCADA data, characterized in that: The following steps are involved: S1, collects the variable pitch system operation data of the wind turbine SCADA system and cleans the operation data; S2, using the local abnormality factor algorithm to monitor the operating data after cleaning in real time to determine whether there is an abnormality in the pitch system. If the pitch system is abnormal, execute S3; if the pitch system is not abnormal, execute S1 again; S3, extract the feature vector of abnormal operation data, input the feature vector into the fault diagnosis model based on deep learning, and diagnose and locate the fault.
2. A method for locating a pitch system fault based on wind turbine SCADA data according to claim 1, characterized in that: The operation data is cleaned in S1, specifically including: The running data are standardized and missing value processed in turn to obtain normal data and abnormal data. A support vector single classification data cleaning model is constructed based on the normal data. The support vector single classification data cleaning model is trained based on the normal data and abnormal data. The trained support vector single classification data cleaning model is used to clean the running data.
3. A method for locating a pitch system fault based on wind turbine SCADA data according to claim 2, characterized in that: The standardization process specifically includes: scaling the operating data to a uniform scale range; The missing value processing specifically includes: using a time series-based interpolation algorithm to fill the missing values of the operating data after standardization processing, marking the operating data whose missing values do not exceed the preset threshold as normal data, and marking the operating data whose missing values exceed the preset threshold as abnormal data.
4. The method for locating a pitch system fault based on wind turbine SCADA data according to claim 2, characterized in that: The support vector single classification data cleaning model is trained based on normal data and abnormal data, and the running data is cleaned using the trained support vector single classification data cleaning model, specifically including: Normal data and abnormal data are used as prepared data, the prepared data are divided into a training set and a test set, various parameters of the support vector single classification data cleaning model are set, the support vector single classification data cleaning model is trained using the training set, and then various indicators of the support vector single classification data cleaning model are calculated using the test set. The trained support vector single classification data cleaning model is evaluated according to the various indicators, and then the parameters and data volume of the support vector single classification data cleaning model are tuned according to the evaluation results to obtain a trained support vector single classification data cleaning model; The operating data is input into the trained support vector single classification data cleaning model, and the operating data is identified by the support vector single classification data cleaning model. If it is identified as normal data, the normal data is directly output; if it is identified as abnormal data, the abnormal data is cleaned, where the cleaning includes: using a smoothing correction algorithm based on adjacent data for cleaning, or deleting the abnormal data.
5. The method for locating a pitch system fault based on wind turbine SCADA data according to claim 1, characterized in that: In S2, a local anomaly factor algorithm is used to monitor the cleaned operation data in real time, specifically including: The local abnormality factor algorithm is used to calculate the local abnormality factor score in the cleaned operating data. The abnormality threshold is set according to the local abnormality factor score. When the local abnormality factor score is greater than or equal to the abnormality threshold, it is determined that there is an abnormality in the variable pitch system; when the local abnormality factor score is less than the abnormality threshold, it is determined that there is no abnormality in the variable pitch system.
6. The method for locating a fault in a variable pitch system of a wind turbine based on SCADA data according to claim 1, characterized in that: The feature vectors of the abnormal operation data in S3 include: time domain feature vectors, frequency domain feature vectors and time-frequency domain feature vectors.
7. The method for locating a pitch system fault based on wind turbine SCADA data according to claim 1, characterized in that: The fault is diagnosed and located in S3, which specifically includes: If a fault is diagnosed in the pitch system, a fault tree model is established based on the fault type, combined with the structure and working principle of the pitch system. The fault tree model uses the fault type as the top event, analyzes the intermediate events and bottom events that lead to the fault, and determines the specific location of the fault by calculating the probability and importance of the bottom event. If it is diagnosed that there is no fault in the pitch system, S1 is executed again.
8. A wind turbine pitch system fault location system based on SCADA data, characterized in that: include: Data acquisition and cleaning module, used to collect the variable pitch system operating data of the wind turbine SCADA system and clean the operating data; The data monitoring module is used to monitor the operating data after cleaning in real time using a local abnormality factor algorithm to determine whether there is any abnormality in the pitch system; The fault diagnosis and location module is used to extract the feature vectors of abnormal operation data, input the feature vectors into the fault diagnosis model based on deep learning, and diagnose and locate the fault.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.