Wind turbine fault diagnosis and early warning method, device, equipment and medium

Abstract

The present application relates to the field of wind power generation technology, and discloses a wind turbine fault diagnosis and early warning method, device, equipment and medium, the method comprises the following steps: obtaining various types of faults corresponding to the physical characteristics generated by the diagnosis object and the working condition information associated with the physical characteristics; the obtained physical characteristics and working condition information are identified and classified, and the principal component characteristics, auxiliary characteristics, and target working condition information associated with the principal component characteristics and auxiliary characteristics are obtained; the principal component characteristics and auxiliary characteristics, target working condition information and single feature threshold value set by the user are fused and analyzed; the alarm conclusion of the diagnosis object obtained after the analysis and processing is output and displayed. The wind turbine fault diagnosis and early warning method can adapt to the habits of users, automatically analyze and diagnose the fault of the diagnosis object for different users, has multi-feature fusion analysis capability, improves the fault analysis efficiency and fault diagnosis and early warning accuracy.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, and in particular to a method, device, equipment and medium for fault diagnosis and early warning of wind turbine generators. Background Technology

[0002] Conventional manual diagnostic analysis methods struggle to provide accurate fault diagnosis conclusions, while current online monitoring and fault diagnosis methods are overly simplistic and unsuitable for diverse user habits. For turbine manufacturers, their "Internet+" platforms lack fundamental data support, resulting in low accuracy in fault diagnosis and early warning, and limited ability to optimize turbine design based on operational feedback. For wind farm operators, significant differences between turbine models and systems make it difficult to scientifically assess the merits of each unit, and data access is challenging. For component suppliers, insufficient data support regarding product reliability prevents them from obtaining accurate component failure rates and primary failure modes. Furthermore, the unique nature of the wind power industry leads to substantial after-sales technical support and communication costs, along with excessively long response times, resulting in low maintenance efficiency. In addition, some system developers are limited by their core algorithm models, making it difficult to guarantee data quality. This poses significant challenges to user-side data analysis, and the data analysis software at the wind farm level is functionally limited, resulting in low data analysis efficiency and hindering quantitative and qualitative analysis of fault phenomena.

[0003] Therefore, how to automatically analyze and diagnose the fault phenomena of mechanical components based on different user needs is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, equipment, and medium for fault diagnosis and early warning of wind turbine units, which can be applied to different users to automatically analyze the diagnostic objects and provide fault diagnosis and early warning, and has the ability to perform multi-feature fusion analysis.

[0005] To address the aforementioned technical problems, this invention provides a method for fault diagnosis and early warning of wind turbine generators, the method comprising: Obtain the physical characteristics corresponding to various faults generated by the diagnostic object, as well as the operating condition information associated with the physical characteristics; The acquired physical features and operating condition information are subjected to outlier identification and classification to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features. The principal and auxiliary features, target operating condition information, and custom-set single feature thresholds are fused and analyzed. The alarm conclusions of the diagnostic object obtained after analysis and processing will be output and displayed.

[0006] In a first aspect, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after obtaining the principal component features, auxiliary features, and target operating condition information associated with the principal component features and auxiliary features, the method further includes: The trend analysis interface displays all principal and auxiliary features; Receives the principal and auxiliary features that are custom-combined through the trend analysis interface, as well as the single feature threshold that is custom-set.

[0007] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, before receiving the principal component features and auxiliary features customized by the trend analysis interface, the method further includes: Determine whether to modify the feature trend combination analysis configuration; If not, load the default feature trend combination analysis configuration information; If so, proceed to the feature value filtering and configuration stage, receive the principal and auxiliary features customized through the trend analysis interface, modify the trend combination analysis configuration and update it synchronously, and load the modified feature trend combination analysis configuration information.

[0008] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after loading the feature trend combination analysis configuration, it further includes: Based on the loaded characteristic trend combination analysis configuration information, relevant characteristic trend data queries are performed; The feature trend data corresponding to the primary and auxiliary features retrieved are combined, plotted, and displayed.

[0009] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the fusion analysis of principal features and auxiliary features, target operating condition information and custom-set single feature thresholds includes: Load airborne system configuration information; Load comprehensive decision parameter information; the comprehensive decision parameter information includes the pre-alarm thresholds and weight information of the custom-set principal features and auxiliary features; The airborne system configuration information, the comprehensive decision parameter information, the principal and auxiliary features, and the target operating condition information are fused and analyzed.

[0010] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after loading the onboard system configuration information and before loading the comprehensive decision parameter information, it further includes: Determine whether to configure custom comprehensive decision parameters; If not, load the default comprehensive decision parameter information; If so, grant permission to configure comprehensive decision parameters, receive comprehensive decision parameter information modified through the comprehensive decision parameter configuration page, and load the modified comprehensive decision parameter information.

[0011] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the step of fusing and analyzing the airborne system configuration information, the comprehensive decision parameter information, the principal component features and auxiliary features, and the target operating condition information includes: Obtain the feature trend charts corresponding to the principal and auxiliary features after the combined drawing; The airborne system configuration information, the comprehensive decision parameter information, the feature trend chart, and the target operating condition information associated with the feature trend chart are input into the comprehensive decision analysis algorithm model for fusion analysis and processing.

[0012] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the process of fusion analysis processing of the comprehensive decision analysis algorithm model includes: Anomaly identification and processing are performed on the feature trend charts and historical feature trend data corresponding to the input principal features and auxiliary features to obtain the processed mutable and non-mutable features. Identify the input target operating condition information; Based on the airborne system configuration information, the pre-alarm threshold and weight information in the comprehensive decision parameter information, the severity of the processed mutable features and the identified target operating condition information is calculated by weighting, and the severity of the fault is obtained, and an alarm conclusion is output.

[0013] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the step of obtaining the physical characteristics corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics includes: Determine all physical characteristics of various faults of the diagnostic object from their occurrence to their failure, as well as the operating condition information associated with these physical characteristics; Obtain the physical characteristics that the diagnostic object can detect, as well as the operating condition information associated with the physical characteristics, from the sensor, SCADA system, or CMS system.

[0014] On the other hand, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the diagnostic object is any one or any combination of multiple components such as transmission chain, tower, tower base, impeller, nacelle, blade, hub, main shaft, and high-strength bolts of tower.

[0015] On the other hand, the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention further includes: Based on the trend patterns of the primary and auxiliary features and the alarm conclusions, a health assessment level and corresponding operation and maintenance recommendations are generated for the diagnostic object.

[0016] To address the aforementioned technical problems, the present invention also provides a wind turbine fault diagnosis and early warning device, the device comprising: The feature acquisition module is used to acquire the physical features corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical features; The feature classification module is used to identify and classify outliers of the acquired physical features and operating condition information to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features. The analysis and processing module is used to perform fusion analysis and processing on principal and auxiliary features, target working condition information, and custom-set single feature thresholds. The alarm output module is used to output and display the alarm conclusions of the diagnostic object obtained after analysis and processing.

[0017] To address the aforementioned technical problems, the present invention also provides a wind turbine fault diagnosis and early warning device, the device comprising: Memory, used to store computer programs; A processor is used to implement the steps of the above-described wind turbine fault diagnosis and early warning method when executing the computer program.

[0018] To address the aforementioned technical problems, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned wind turbine fault diagnosis and early warning method.

[0019] As can be seen from the above technical solution, the wind turbine fault diagnosis and early warning method provided by the present invention includes: acquiring the physical characteristics corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics; performing outlier identification and classification on the acquired physical characteristics and operating condition information to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features; performing fusion analysis on the principal features and auxiliary features, target operating condition information, and a custom-set single feature threshold; and outputting and displaying the alarm conclusion of the diagnostic object obtained after analysis and processing.

[0020] The beneficial effects of this invention are that, by using the wind turbine fault diagnosis and early warning method provided by this invention, the detectable physical characteristics of the diagnostic object and the associated operating condition information are first obtained. After corresponding processing, the principal component features, auxiliary features, and associated target operating condition information are obtained. Then, the feature data and operating condition information obtained by combining user-defined single feature thresholds are fused and analyzed to output alarm conclusions. This can adapt to user habits, ensure data quality, and is suitable for different users to automatically analyze the diagnostic object and perform fault diagnosis and early warning. It has the ability to perform multi-feature fusion analysis, improves fault analysis efficiency and fault diagnosis and early warning accuracy.

[0021] In addition, the present invention also provides a corresponding wind turbine fault diagnosis and early warning device, wind turbine fault diagnosis and early warning equipment and computer-readable storage medium for the wind turbine fault diagnosis and early warning method, which have the same or corresponding technical features as the wind turbine fault diagnosis and early warning method mentioned above, and have the same effect. Attached Figure Description

[0022] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart of a wind turbine fault diagnosis and early warning method provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the wind turbine fault diagnosis and early warning method provided in this embodiment of the invention. Figure 3 A flowchart of the feature trend combination analysis process provided in this embodiment of the invention; Figure 4 A flowchart of the comprehensive decision-making process provided in the embodiments of the present invention; Figure 5 A flowchart of the data fusion and analysis process provided in this embodiment of the invention; Figure 6 This is a schematic diagram of the structure of the wind turbine fault diagnosis and early warning device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of a wind turbine fault diagnosis and early warning device provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present invention.

[0025] Different users, including specialists and diagnostic analysts, have different focuses and usage habits regarding the fault characteristics, trends, and early warnings of wind turbine mechanical components. The core of this invention is to provide a method, device, equipment, and medium for fault diagnosis and early warning of wind turbine components, addressing the technical problem that current fault diagnosis methods for wind turbine mechanical components are inefficient and cannot be adapted to the different user habits.

[0026] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Figure 1 A flowchart of a wind turbine fault diagnosis and early warning method provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes: S101. Obtain the physical characteristics of various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics.

[0027] It should be noted that the diagnostic object refers to the specific diagnostic component among the major mechanical components of the wind turbine. The diagnostic object may include, but is not limited to, any one or any combination of the following: drivetrain (such as drivetrain main shaft bearings and shaft system, gearbox bearings, gears), tower, tower base, impeller, nacelle, blades, hub, main shaft, and high-strength bolts of the tower.

[0028] Various types of faults can be categorized into typical faults and system faults. Typical faults are those caused by inherent defects in the turbine itself, such as bearing outer ring failures or gear pitting. System faults are phenomena such as abnormal turbine performance, excessive vibration, and abnormal noise caused by problems in the design, assembly, installation, commissioning, and operation and maintenance of the wind turbine. Examples include component damage or shutdown protection due to poor component matching design, damage to other components due to improper or untimely maintenance and repair, and component damage caused by the transmission of external stress (electrical) factors.

[0029] The physical characteristics of major mechanical components in wind turbines vary significantly between different types of diagnostic objects. Diagnostic methods are based on applied statistical data, expert experience, and the fault-causing mechanisms of the diagnostic objects to determine the physical quantities generated from the onset of a fault to its final failure. For example, the physical characteristics generated by diagnostic objects such as drivetrain bearings, gears, and shafts include, but are not limited to, vibration, impact, temperature, noise, oil abrasive particles, axial movement, misalignment, imbalance, and shaft voltage. The physical characteristics generated by diagnostic objects such as the tower include, but are not limited to, vibration acceleration, sway amplitude, tilt angle and direction, and modal frequency changes; uneven settlement of the tower foundation; impeller dynamic imbalance; nacelle vibration acceleration and sway; blade vibration acceleration and deformation, and blade modal frequency changes; and loosening or breakage of high-strength bolts.

[0030] In specific implementation, step S101 acquires the physical characteristics corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics. Specifically, it may include: first, determining all physical characteristics generated by various faults of the diagnostic object from their occurrence to their failure and the operating condition information associated with the physical characteristics; then, acquiring the physical characteristics that the diagnostic object can detect and the operating condition information associated with the physical characteristics from sensors or supervisory control and data acquisition (SCADA) or condition monitoring system (CMS).

[0031] In implementation, such as Figure 2 As shown, based on application statistics, expert experience, and the fault generation mechanism of the diagnostic object, all physical characteristics from the occurrence of the fault to its failure, along with associated operating condition information, are determined. Then, corresponding sensors are installed to collect and extract, or obtain from third-party systems (such as SCADA systems or CMS systems), the physical characteristics that the diagnostic object can detect, along with associated operating condition information.

[0032] CMS can monitor vibration signals and extract component state characteristics through data processing in different dimensions such as time domain, frequency domain, and time-frequency domain. Analysis based on time domain statistical characteristics is the most commonly used monitoring method in time domain analysis. Commonly used monitoring indicators include mean, peak value, root square value, root mean square value, variance, slope, and kurtosis. SCADA can monitor status information such as temperature and control feedback information.

[0033] Vibration trend characteristic indicators can quantify the vibration data of massive wind turbine drive systems, intuitively displaying the status information of the wind turbine drive system contained in the data. Their changing trends over time can reflect the changing process of the wind turbine drive system's operating status. Setting reasonable thresholds for these characteristic indicators can provide timely alarms when the wind turbine drive system exhibits abnormal conditions. Traditional vibration analysis mainly focuses on whether the effective value of acceleration and intensity values ​​exceed international and similar national standards (VDI3834, GBT35854-2018 standards) to determine the health status of various wind turbine components. This method can only identify vibration anomalies in large components of the wind turbine and requires analysts to have certain vibration analysis knowledge, placing certain demands on the quantity and quality of human resources.

[0034] S102. Perform outlier identification and classification on the acquired physical features and operating condition information to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features.

[0035] It should be noted that principal features refer to the feature values ​​that play a leading role in diagnosis and health assessment. Auxiliary features refer to the feature values ​​that play a supporting or confirmatory role in diagnosis and health assessment.

[0036] After processing, the physical characteristics generated by the diagnostic objects of the mechanical major components of the wind turbine can be used to extract the corresponding principal and auxiliary features, and simultaneously obtain the associated operating condition information, such as the dB value and temperature value of the bearing principal feature, the abrasive particle value of the auxiliary feature, the misalignment and eccentricity of the shaft system, and the effective vibration value of the bearing location; the operating condition information includes but is not limited to wind speed, power generation, rotational speed, and ambient temperature.

[0037] In implementation, step S102 can perform secondary processing, outlier identification, and classification on the detectable physical features and operating condition information of the diagnostic object to obtain accurate and comprehensive principal component features, auxiliary features, and operating condition information associated with the principal component features and auxiliary features. This step can be executed through the data acquisition, processing, and feature extraction module and the operating condition data input module.

[0038] After obtaining the principal features, auxiliary features, and target working condition information associated with the principal features and auxiliary features, such as... Figure 2 As shown, this can include storing all the principal features, auxiliary features, and target operating condition information of the diagnostic object into the database of the online fault early warning and intelligent operation and maintenance platform software. The online fault early warning and intelligent operation and maintenance platform software is based on a B / S architecture design. Users can access this platform to achieve functions such as online status monitoring of wind turbine mechanical components, feature trend combination analysis, sample analysis, automatic fault diagnosis, health assessment, operation and maintenance management, and system settings.

[0039] S103. Perform fusion analysis on the principal features and auxiliary features, target working condition information, and custom-set single feature thresholds.

[0040] It should be noted that the single feature threshold refers to the pre-alarm threshold for each feature.

[0041] In implementation, step S103 allows users, as specialists or diagnostic analysis engineers, to customize single-feature thresholds for major mechanical components of wind turbines based on standards such as VDI3834 and GBT35854-2018 on the trend analysis interface of the online fault early warning and intelligent operation and maintenance platform software (hereinafter referred to as the trend analysis interface). Based on the principal features, auxiliary features, associated operating conditions, and single-feature thresholds set by the user in the trend analysis interface, a combined weighted statistical decision is made to automatically diagnose and output pre-alarms, thereby achieving automatic diagnosis of the diagnostic object. Automatic diagnosis is a method of automatically determining the presence, location, and magnitude of a fault. This step can be executed through the integrated decision module.

[0042] In practical applications, when users do not have personalized needs or are not yet proficient, the design of a standard trend analysis interface for the diagnostic object and an automatic diagnostic output and pre-alarm mode based on the combination of principal and auxiliary features for weighted statistical comprehensive decision-making can be used to achieve automatic diagnosis of the diagnostic object.

[0043] S104. Output and display the alarm conclusions of the diagnostic object obtained after analysis and processing.

[0044] In the wind turbine fault diagnosis and early warning method provided in this embodiment of the invention, the detectable physical characteristics of the diagnostic object and the associated operating condition information are first obtained. After corresponding processing, the principal features, auxiliary features, and associated target operating condition information are obtained. Then, the feature data and operating condition information obtained by combining user-defined single feature thresholds are fused and analyzed to output an alarm conclusion. This can adapt to user habits, ensure data quality, and is suitable for different users to perform automatic analysis and fault diagnosis and early warning of the diagnostic object. It has the ability to perform multi-feature fusion analysis, improve the efficiency of fault analysis and the accuracy of fault diagnosis and early warning.

[0045] It should be noted that the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention can serve the wind turbine online fault early warning and intelligent operation and maintenance system. The system adopts a 1+X+N platform architecture and is supported by the backend.

[0046] Among them, "1" refers to the platform software, which may include a smart wind farm information support sub-platform. This sub-platform consists of a background expert diagnostic system and artificial intelligence self-learning software deployed at the central control or centralized control terminal of the smart wind farm to realize online fault early warning, health assessment, and smart operation and maintenance services. Its functions include: realizing component fault early warning, health assessment and maintenance spare parts, and order placement suggestions based on multi-information fusion and comprehensive decision-making; providing operation and maintenance closed-loop platform support; and intelligently optimizing automatic diagnostic conclusions and operation and maintenance suggestion library based on user habits. It also reserves an interface with the wind turbine main control system and can achieve data interconnection with the enterprise operation and maintenance platform, providing standard-defined status and conclusion data to the full-scale warehouse of big data industrial internet.

[0047] X refers to the subsystem, which may include a transmission chain monitoring subsystem, a tower overturning monitoring subsystem, a tower base monitoring subsystem, an impeller (nacelle) monitoring subsystem, a blade monitoring subsystem, a high-strength bolt monitoring subsystem, a SCADA data acquisition subsystem, a CMS data processing subsystem, etc. It can realize the data acquisition, feature extraction and data filtering of various mechanical components for diagnosis, and send the data to the wind farm or centralized control platform server. Each subsystem can be expanded and connected as needed.

[0048] N refers to N terminals, which can include App (mini-program) operation and maintenance terminals, monitoring PC terminals, enterprise operation and maintenance platform large screen terminals, etc., to realize the closed-loop implementation of monitoring and operation and maintenance, or can be achieved by utilizing the user's existing terminals.

[0049] In addition, backend support can include support for enterprise big data center data mining and fault mechanism diagnosis; provide online intentional early warning and intelligent operation and maintenance services for monitored and diagnosed objects; and provide an authorized and certified open platform to enable multi-party participation and interconnection.

[0050] For example, the diagnostic object in the drivetrain monitoring subsystem is the drivetrain. For the drivetrain, fault types can include seven types of bearing faults and gear faults. The monitoring content can be based on machine model configuration and speed tracking sampling. Principal features can include bearing dB values ​​(outer ring, inner ring, outer ring, single rolling, double rolling, and the same shaft) and gear dB values ​​(the same tooth and adjacent teeth). When the principal feature is the bearing dB value, the corresponding auxiliary features can include SV value (obtained from SCADA), temperature value (obtained from SCADA), effective acceleration value, effective velocity value, mean, peak-to-peak value, and maximum value, 1st / 2nd order spectral amplitude, 1st order spectral amplitude, 2nd order spectral amplitude, 3rd order spectral amplitude, 4th order spectral amplitude, and 6th order spectral amplitude, plastic steel cage wear, and external arc spectrum dB value. When the primary characteristic is the gear dB value, the corresponding auxiliary characteristics may include SV value, meshing spectrum dB amplitude, temperature value, effective acceleration value, effective velocity value, mean, peak-to-peak value, and maximum value, frequency 1 / 2 order spectral amplitude, frequency 1st order spectral amplitude, frequency 2nd order spectral amplitude, frequency 3rd order spectral amplitude, frequency 4th order spectral amplitude, and frequency 6th order spectral amplitude, meshing spectrum amplitude, and accumulated wear particles (ferromagnetic and non-ferromagnetic) 24-hour accumulation number and size (three intervals) characteristic values ​​(provided when the oil sensor is installed). At this time, the target operating condition information may include wind speed, shaft speed, power, generator speed, and nacelle ambient temperature (if SCADA data cannot be obtained, self-monitored operating condition speed information will be used).

[0051] For example, the diagnostic object in the tower tilting monitoring subsystem is the tower itself. For the tower, when the fault type / monitored content is excessive vibration acceleration at the top or middle of the tower, the main element features may include the effective value of acceleration and the effective value of velocity. Corresponding auxiliary features may include the installation x-direction acceleration, installation y-direction acceleration, installation zero-position angle, and yaw angle (nacelle position angle, obtained from SCADA). When the fault type / monitored content is excessive tilt angle at the top or middle of the tower, the main element features may include the tilt angle and the azimuth angle (which direction it tilts towards, the angle relative to the zero-position angle). Corresponding auxiliary features may include the installation x-direction acceleration, installation y-direction acceleration, installation zero-position angle, and yaw angle (nacelle position angle, obtained from SCADA). When the fault type / monitored content is excessive amplitude at the top or middle of the tower, the main element features may include the sway amplitude in the main shaft direction and the lateral direction of the main shaft (or the maximum displacement if yaw angle information is unavailable). Corresponding auxiliary features may include the installation x-direction acceleration, installation y-direction acceleration, installation zero-position angle, and yaw angle (nacelle position angle, obtained from SCADA). When the fault type / monitored content is multi-mode frequency variation and amplitude exceeding limits at the tower top and middle, the principal component features may include the first and second order modal frequencies and amplitudes at the front and rear of the tower (main axis direction), the first and second order modal frequencies and amplitudes at the side (lateral direction of the main axis), and the first order torsional modal frequency and amplitude. Corresponding auxiliary features may include installation x-axis acceleration, installation y-axis acceleration, installation zero-position angle, yaw angle (nacelle position angle, obtained from SCADA), theoretical values ​​of the first and second order modal frequencies at the front and rear of the tower (main axis direction), theoretical values ​​of the first and second order modal frequencies at the side (lateral direction), and theoretical value of the first order torsional modal frequency. In this case, the target operating condition information may include wind speed, rotor speed, power, generator speed, and nacelle ambient temperature (if SCADA data is unavailable, self-monitored operating condition speed information will be used).

[0052] For example, the diagnostic object in the tower base monitoring subsystem is the tower base. For the tower base, when the fault type / monitoring content is monitoring uneven settlement of the tower base, the principal component features may include the absolute tilt angle of the tower base and the relative tilt angle of the tower base. Corresponding auxiliary features may include the initial installation value and the tower base diameter. When the fault type / monitoring content is monitoring the movement of the tower relative to the tower base, the principal component feature is the tilt angle of the tower wall relative to the tower base. Corresponding auxiliary features may include the absolute tilt angle of the tower wall, the relative tilt angle of the tower wall, and the initial installation value. In this case, the target operating condition information may include wind speed, rotor speed, power, generator speed, and nacelle ambient temperature.

[0053] For example, the diagnostic objects in the impeller (nacelle) monitoring subsystem may include the impeller and the nacelle. For the impeller, when the fault type / monitoring content is impeller dynamic imbalance monitoring, the principal component features may include the 1st, 2nd, and 3rd order spectral amplitudes of the main shaft direction rotational frequency and the 1st, 2nd, and 3rd order spectral amplitudes of the main shaft transverse rotational frequency. Corresponding auxiliary features may include the RMS vibration acceleration before and after the vibration protection switch nacelle (obtained from SCADA) and the RMS vibration acceleration on the left and right sides of the vibration protection switch nacelle (obtained from SCADA). When the fault type / monitoring content is blade passing frequency monitoring, the principal component features may include the 1st, 2nd, and 3rd order spectral amplitudes of the blade passing frequency in the main shaft direction and the 1st, 2nd, and 3rd order spectral amplitudes of the blade passing frequency in the main shaft transverse direction. Corresponding auxiliary features may include the RMS vibration acceleration before and after the vibration protection switch nacelle (obtained from SCADA) and the RMS vibration acceleration on the left and right sides of the vibration protection switch nacelle (obtained from SCADA). For the nacelle, when the fault type / monitoring content is nacelle vibration status monitoring, the principal component features may include the effective value of the main shaft acceleration, the effective value of the velocity, and the effective value of the main shaft lateral acceleration. Corresponding auxiliary features may include the RMS vibration acceleration aft and rear of the nacelle (obtained from SCADA) and the RMS vibration acceleration left and right of the nacelle (obtained from SCADA). In this case, the target operating condition information may include wind speed, rotor speed, power, generator speed, and nacelle ambient temperature.

[0054] For example, the diagnostic object in the blade monitoring subsystem is the blade. For the blade, when the fault type / monitoring content is blade vibration (deformation, imbalance) monitoring, the principal component features may include the effective value of the blade tip acceleration, the effective value of the blade tip lateral acceleration, and the effective value of the acceleration perpendicular to the blade cross section. Corresponding auxiliary features may include the pitch angle of each blade (obtained from SCADA) and the temperature of the pitch motor of each blade (obtained from SCADA). When the fault type / monitoring content is blade structural damage (leading and trailing edge cracking, blade root cracking, blade surface damage, severe lightning strike damage), the principal component features may include the blade oscillation mode frequency and amplitude, and the blade flapping mode frequency and amplitude. Corresponding auxiliary features may include the pitch angle of each blade (obtained from SCADA) and the temperature of the pitch motor of each blade (obtained from SCADA). When the fault type / monitoring content is icing monitoring, the principal component features may include the blade oscillation mode frequency and amplitude, and the blade flapping mode frequency and amplitude. Corresponding auxiliary features may include hub temperature and humidity, and the temperature of the pitch motor of each blade (obtained from SCADA). At this point, the target operating condition information may include wind speed, wind turbine speed, power, generator speed, and nacelle ambient temperature.

[0055] For example, the diagnostic targets in the high-strength bolt monitoring subsystem are high-strength bolts in the hub, main shaft, and tower. For these high-strength bolts, when the fault type / monitoring content is high-strength bolt loosening or breakage, the primary feature may include ultrasonic warning force, displacement change between flanges, and the relative loosening angle of the bolt; there are no corresponding auxiliary features. In this case, the target operating condition information may include wind speed, rotor speed, power, generator speed, and nacelle ambient temperature.

[0056] Furthermore, in specific implementations, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, such as... Figure 2 As shown, it may also include: generating a health assessment level of the diagnostic object and corresponding operation and maintenance suggestions based on the trend patterns of the principal features and auxiliary features, as well as the alarm conclusions.

[0057] During implementation, the health assessment is conducted based on the trends and patterns of primary and auxiliary characteristics obtained from the physical phenomena resulting from the faults of the diagnosed object, as well as the frequency of early warnings output by the comprehensive decision-making process. This assessment provides the health level of the diagnosed object and corresponding maintenance recommendations. This step can be performed through the health assessment module.

[0058] Furthermore, in a specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after obtaining the principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features in step S102, the method may further include: displaying all principal features and auxiliary features using a trend analysis interface; receiving principal features and auxiliary features that are custom-combined through the trend analysis interface, as well as custom-set single feature thresholds.

[0059] In implementation, the online fault early warning and intelligent operation and maintenance platform software combines trend analysis interfaces to display customized trend combinations of diagnostic features of major mechanical components of wind turbines, based on the user's preferences or the diagnostic analysis engineer's own habits and areas of interest. The platform software has a memory function, allowing user engineers or diagnostic analysis engineers to set pre-alarm thresholds for key features of interest, and the data or files are saved in the background. The next time the software is accessed, the customized trend analysis interface will be displayed.

[0060] In specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after displaying all primary and secondary features using the trend analysis interface, before receiving the primary and secondary features and auxiliary features customized through the trend analysis interface, the method may further include: determining whether to modify the feature trend combination analysis configuration; if not, loading the default feature trend combination analysis configuration information; if yes, entering the feature value filtering configuration stage, receiving the primary and secondary features and auxiliary features customized through the trend analysis interface to modify the trend combination analysis configuration and update it synchronously, and loading the modified feature trend combination analysis configuration information.

[0061] In implementation, such as Figure 3 As shown, the system determines whether to modify the trend combination analysis configuration. If no configuration is made, the default feature trend combination analysis configuration information is loaded. If configuration is made, the system enters the feature value filtering configuration stage, where users can modify the trend combination analysis configuration information and update it synchronously to the database. In this stage, the relevant feature values ​​of the diagnostic object are filtered, and the filtered configuration information is synchronously updated to the database. Finally, the system queries and loads the trend combination analysis configuration information from the database.

[0062] It should be noted that this invention is not a fixed feature trend early warning mode. Based on the research of the fault mechanism of the diagnostic object, more features are extracted from the data collected at the edge. Standard mode and user configuration mode are set. This can meet the needs of users without personalized needs, as well as the needs of user specialists or diagnostic analysis engineers for personalized feature trend combination analysis and automatic diagnosis early warning based on their own experience and the actual operation and maintenance situation of the enterprise.

[0063] In specific implementation, the wind turbine fault diagnosis and early warning method provided in the above embodiments of the present invention may further include, after loading the feature trend combination analysis configuration, performing relevant feature trend data query based on the loaded feature trend combination analysis configuration information; and combining, drawing and displaying the feature trend data corresponding to the queried principal features and auxiliary features.

[0064] In implementation, such as Figure 3 As shown, relevant feature trend data is queried based on the trend combination analysis configuration information; the queried trend data is combined into a chart to obtain a feature trend chart, which users can view by cursor, zoom in and out.

[0065] This invention can be applied to different users who use feature trend data to perform analysis and automatic diagnosis and early warning based on their own habits and concerns.

[0066] Furthermore, in a specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, step S103 performs fusion analysis processing on the principal component features and auxiliary features, target operating condition information, and custom-set single feature thresholds. Specifically, it may include: first, loading the airborne system configuration information; then, loading the comprehensive decision parameter information; the comprehensive decision parameter information includes the pre-alarm thresholds and weight information of the custom-set principal component features and auxiliary features; finally, performing fusion analysis processing on the airborne system configuration information, comprehensive decision parameter information, principal component features and auxiliary features, and target operating condition information.

[0067] It should be noted that the airborne system is a collective term for the monitoring and diagnostic subsystems of various large mechanical components installed on the wind turbine. It can collect fault characteristic data of the wind turbine's large mechanical components, extract features, screen raw data, and transmit the data to the ground server.

[0068] In implementation, such as Figure 4 As shown, querying and loading airborne system configuration information can include aircraft type, airborne major component model parameter information, existence identifiers of each airborne subsystem, sensor measurement point configuration information, etc.; then, loading the principal components, auxiliary features, alarm threshold values ​​and weight information of the diagnostic object.

[0069] Furthermore, in a specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, after loading the onboard system configuration information and before loading the comprehensive decision parameter information, it may further include: determining whether to perform custom comprehensive decision parameter configuration; if not, loading the default comprehensive decision parameter information; if so, opening the comprehensive decision parameter configuration permission, receiving the comprehensive decision parameter information modified through the comprehensive decision parameter configuration page, and loading the modified comprehensive decision parameter information.

[0070] In implementation, such as Figure 4 As shown, after loading the airborne system configuration information, it can be determined whether the user has modified the configuration parameters of the diagnostic object's primary element, auxiliary feature pre-alarm threshold, and weight factor. If not configured, the default diagnostic object's primary element, auxiliary feature pre-alarm threshold, and weight parameter information is loaded. If configured, the user is granted permission to configure the comprehensive decision parameters. The user can modify the diagnostic object's primary element, auxiliary feature value pre-alarm threshold, and related weights on the comprehensive decision parameter configuration page according to the fault characteristics they are interested in, and save them synchronously to the database. Then, the modified diagnostic object's primary element, auxiliary feature pre-alarm threshold, and weight parameter information is loaded.

[0071] Furthermore, in specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the airborne system configuration information, comprehensive decision parameter information, principal component features and auxiliary features, and target operating condition information are fused and analyzed. Specifically, this may include: obtaining feature trend charts corresponding to the combined principal component features and auxiliary features; and inputting the airborne system configuration information, comprehensive decision parameter information, feature trend charts, and target operating condition information associated with the feature trend charts into the comprehensive decision analysis algorithm model for fusion and analysis.

[0072] During implementation, the sample characteristic trend data, emergency alarm signs, and operating condition information of the diagnostic object can be queried and cached in the analysis area; the airborne system configuration information, comprehensive decision parameter information, sample characteristic trend data, emergency alarm signs, and operating condition information are sent to the comprehensive decision analysis algorithm model for fusion analysis and processing.

[0073] Furthermore, in specific implementation, in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the process of fusion analysis and processing of the comprehensive decision analysis algorithm model may specifically include: anomaly identification and processing of the feature trend charts and historical feature trend data corresponding to the input principal features and auxiliary features to obtain the processed mutable and non-mutable features; identification of the input target operating condition information; and weighted severity calculation of the processed mutable features and the identified target operating condition information based on the airborne system configuration information, the pre-alarm threshold and weight information in the comprehensive decision parameter information to obtain the severity of the fault and output an alarm conclusion.

[0074] It should be noted that in this invention, anomaly identification and processing, such as cleaning and removal, can be performed on feature data, including feature trend charts and historical feature trend data corresponding to primary and auxiliary features. When feature data jumps significantly within the standard (normal) range but does not conform to the actual value of the operating physical quantity, the feature data is a mutation-prone feature. When feature data exceeds the maximum and minimum ranges but is an outlier of a fixed or slowly changing quantity, the feature data is a non-mutation-prone feature.

[0075] In implementation, such as Figure 5 As shown, based on the latest N (externally configured) principal features and selected auxiliary features, and the historical trend feature data (including principal features and auxiliary features) of the current time and the most recent N-1 times, a total of N data points are used as input. Anomaly identification and processing are performed on the input feature data, where anomaly processing is divided into mutable features and non-mutable features. The input target operating condition information is identified. The anomaly-processed feature values ​​and the identified target operating condition information are combined with the fault prediction and alarm thresholds (where the fault prediction and alarm thresholds need to be externally input and are initially set with default values) to calculate the weighted severity of the fault, and the alarm conclusion is output. The alarm conclusion is divided into two levels: warning and alarm.

[0076] It should be noted that in the wind turbine fault diagnosis and early warning method provided in the embodiments of the present invention, the features of various fault information generated by the diagnostic object can be accurately extracted at the upper edge of the wind turbine. The feature information of the diagnostic object can be collected and classified in the platform software background, such as into primary features and auxiliary features. The present invention provides a trend analysis interface and a pre-alarm mode based on the combined weighted statistical decision-making of primary and auxiliary features. The trend analysis interface allows users or diagnostic engineers to select combinations of features and trends of wind turbine mechanical components that they are accustomed to and interested in, display and analyze these trends, and has a memory function. Users or diagnostic engineers can set pre-alarm thresholds for the primary features they are interested in and save them in the background. The next time the interface is opened, it will be a customized trend analysis interface to suit the diagnostic analysis needs of different users with different feature trend data. Furthermore, users or diagnostic engineers can set corresponding primary and auxiliary features for the diagnostic object to perform combined weighted statistical decision-making and pre-alarm output, to suit the pre-alarm needs of different users based on their own habits and points of interest.

[0077] In the above embodiments, the wind turbine fault diagnosis and early warning method has been described in detail. This invention also provides embodiments of wind turbine fault diagnosis and early warning devices and equipment. It should be noted that this invention describes the embodiments of the device from two perspectives: one based on functional modules, and the other based on hardware.

[0078] Figure 6 This is a structural diagram of a wind turbine fault diagnosis and early warning device according to an embodiment of the present invention. Based on the functional modules, this embodiment includes: The feature acquisition module 10 is used to acquire the physical features corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical features; The feature classification module 11 is used to identify and classify outliers of the acquired physical features and working condition information to obtain principal features, auxiliary features, and target working condition information associated with the principal features and auxiliary features. Analysis and processing module 12 is used to perform fusion analysis and processing on principal component features and auxiliary features, target working condition information and custom-set single feature thresholds; The alarm output module 13 is used to output and display the alarm conclusions of the diagnostic object obtained after analysis and processing.

[0079] In the wind turbine fault diagnosis and early warning device provided in the embodiments of the present invention, the interaction of the above four modules can adapt to user habits, ensure data quality, and be suitable for different users to perform automatic analysis and fault diagnosis and early warning of the diagnostic object. It has the ability to perform multi-feature fusion analysis, thereby improving the efficiency of fault analysis and the accuracy of fault diagnosis and early warning.

[0080] Since the embodiments of the device section correspond to the embodiments of the method section, please refer to the description of the embodiments of the method section for the embodiments of the device section, and they will not be repeated here. Furthermore, it has the same beneficial effects as the wind turbine fault diagnosis and early warning method mentioned above.

[0081] Figure 7 This is a structural diagram of a wind turbine fault diagnosis and early warning device provided in another embodiment of the present invention. This embodiment is based on a hardware perspective, such as... Figure 7 As shown, the wind turbine fault diagnosis and early warning equipment includes: Memory 20 is used to store computer programs; The processor 21 is used to execute a computer program to implement the steps of the wind turbine fault diagnosis and early warning method mentioned in the above embodiments.

[0082] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as the CPU, is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an Artificial Intelligence (AI) processor, which is used to handle computational operations related to machine learning.

[0083] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 20 is used to store at least the following computer program 201, which, after being loaded and executed by the processor 21, can implement the relevant steps of the wind turbine fault diagnosis and early warning method disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, and the storage method may be temporary storage or permanent storage. The operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, the data involved in the aforementioned wind turbine fault diagnosis and early warning method.

[0084] In some embodiments, the wind turbine fault diagnosis and early warning device may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0085] Those skilled in the art will understand that Figure 7 The structure shown does not constitute a limitation on wind turbine fault diagnosis and early warning equipment, and may include more or fewer components than shown.

[0086] The wind turbine fault diagnosis and early warning device provided in this embodiment of the invention includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the following method: wind turbine fault diagnosis and early warning method, with the same effect as above.

[0087] Finally, the present invention also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps described in the above method embodiments.

[0088] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] The computer-readable storage medium provided by this invention includes the aforementioned wind turbine fault diagnosis and early warning method, and has the same effect.

[0090] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only 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 one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0091] The foregoing has provided a detailed description of the wind turbine fault diagnosis and early warning method, apparatus, equipment, and medium provided by this invention. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from the principles of the invention, and these improvements and modifications also fall within the protection scope of the claims of this invention.

Claims

1. A method for fault diagnosis and early warning of wind turbine generators, characterized in that, The method includes: Obtain the physical characteristics of various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics; The acquired physical features and operating condition information are subjected to outlier identification and classification to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features. The trend analysis interface displays all principal and auxiliary features; Determine whether to modify the feature trend combination analysis configuration; if it is determined that the feature trend combination analysis configuration should not be modified, then load the default feature trend combination analysis configuration information; if it is determined that the feature trend combination analysis configuration should be modified, then enter the feature value filtering configuration stage, receive the principal features and auxiliary features customized through the trend analysis interface, modify the trend combination analysis configuration and update it synchronously, and load the modified feature trend combination analysis configuration information. Based on the loaded feature trend combination analysis configuration information, relevant feature trend data are queried, and the feature trend data corresponding to the queried principal features and auxiliary features are combined, drawn, and displayed. Load airborne system configuration information; Determine whether to configure custom comprehensive decision parameters; if it is determined that no custom comprehensive decision parameter configuration is to be performed, load the default comprehensive decision parameter information; if it is determined that custom comprehensive decision parameter configuration is to be performed, grant comprehensive decision parameter configuration permissions, receive comprehensive decision parameter information modified through the comprehensive decision parameter configuration page, and load the modified comprehensive decision parameter information; wherein, the comprehensive decision parameter information includes the pre-alarm thresholds and weight information of the custom-set principal features and auxiliary features; The airborne system configuration information, the comprehensive decision parameter information, the principal component features and auxiliary features, and the target operating condition information are fused and analyzed. The alarm conclusions of the diagnostic object obtained after analysis and processing will be output and displayed.

2. The wind turbine fault diagnosis and early warning method according to claim 1, characterized in that, The process of fusing and analyzing the airborne system configuration information, the integrated decision parameter information, principal component features and auxiliary features, and the target operating condition information includes: Obtain the feature trend charts corresponding to the principal and auxiliary features after the combined drawing; The airborne system configuration information, the comprehensive decision parameter information, the feature trend chart, and the target operating condition information associated with the feature trend chart are input into the comprehensive decision analysis algorithm model for fusion analysis and processing.

3. The wind turbine fault diagnosis and early warning method according to claim 2, characterized in that, The process of fusion analysis processing in the comprehensive decision analysis algorithm model includes: Anomaly identification and processing are performed on the feature trend charts and historical feature trend data corresponding to the input principal features and auxiliary features to obtain the processed mutable and non-mutable features. Identify the input target operating condition information; Based on the airborne system configuration information, the pre-alarm threshold and weight information in the comprehensive decision parameter information, the processed mutable features and the identified target operating condition information are weighted and the severity is calculated to obtain the severity of the fault and output an alarm conclusion.

4. The wind turbine fault diagnosis and early warning method according to claim 1, characterized in that, The acquisition of the physical characteristics corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical characteristics includes: Determine all physical characteristics of various faults of the diagnostic object from their occurrence to their failure, as well as the operating condition information associated with these physical characteristics; Obtain the physical characteristics that the diagnostic object can detect, as well as the operating condition information associated with the physical characteristics, from the sensor, SCADA system, or CMS system.

5. The wind turbine fault diagnosis and early warning method according to claim 1, characterized in that, The diagnostic objects are any one or any combination of several of the following: transmission chain, tower, tower base, impeller, nacelle, blade, hub, main shaft, and high-strength bolts of the tower.

6. The wind turbine fault diagnosis and early warning method according to claim 1, characterized in that, Also includes: Based on the trend patterns of the primary and auxiliary features and the alarm conclusions, a health assessment level and corresponding operation and maintenance recommendations are generated for the diagnostic object.

7. A fault diagnosis and early warning device for wind turbine generators, characterized in that, The device is used to perform the wind turbine fault diagnosis and early warning method as described in any one of claims 1 to 6, the device comprising: The feature acquisition module is used to acquire the physical features corresponding to various faults generated by the diagnostic object and the operating condition information associated with the physical features; The feature classification module is used to identify and classify outliers of the acquired physical features and operating condition information to obtain principal features, auxiliary features, and target operating condition information associated with the principal features and auxiliary features. The analysis and processing module is used to perform fusion analysis and processing on principal and auxiliary features, target working condition information, and custom-set single feature thresholds. The alarm output module is used to output and display the alarm conclusions of the diagnostic object obtained after analysis and processing.

8. A wind turbine fault diagnosis and early warning device, characterized in that, The device includes: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of the wind turbine fault diagnosis and early warning method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the wind turbine fault diagnosis and early warning method as described in any one of claims 1 to 6.

Images