Automatic analysis method for fan electric quantity loss reason based on machine learning

Abstract

The invention belongs to the technical field of wind power generation, and discloses an automatic analysis method and device for the power loss reason of a fan, and the method comprises the following steps: collecting the operation data, environment data and state data of the fan; carrying out segmented modeling on the wind speed range; correcting the air density and the turbulivity data; calculating electric quantity loss and extracting multi-dimensional features, wherein the multi-dimensional features comprise wind regime features, power features, environment features and motion features; and identifying the loss reason of the multi-dimensional feature through an ensemble learning method. According to the method, the problem of mass data processing is effectively solved. And the fitting precision of the power curve is obviously improved. Through multi-dimensional feature extraction and an integrated learning framework, automatic identification of the power loss reason is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of wind power generation, and in particular relates to a method and a device for automatically analyzing the cause of power loss of a wind turbine. Background Art

[0002] As a clean and renewable energy source, wind power generation plays an increasingly important role in the transformation of the global energy structure. During the operation of wind turbines, due to the influence of complex natural environment and equipment operation status, the problem of power generation efficiency loss often occurs. This power loss not only directly affects the economic benefits of wind farms, but also affects the stable operation of the power grid. Therefore, accurately identifying and analyzing the causes of wind turbine power loss is of great significance to improving the operational efficiency and economic benefits of wind farms. At present, the analysis of the causes of wind turbine power loss mainly relies on manual experience judgment or simple statistical analysis methods. The traditional analysis method has the following problems: First, the analysis process is heavily dependent on the experience of operation and maintenance personnel and lacks scientific quantitative analysis methods; second, the analysis method is too simple and cannot handle massive operating data and complex environmental factors; third, the timeliness of the analysis results is poor, and the analysis results are often obtained a long time after the loss occurs, resulting in the inability to take optimization measures in time. These problems seriously restrict the efficiency improvement of wind farms.

[0003] The Chinese patent publication number is CN113623143A, and the name is a patent application for a system for automatically identifying the root cause of power loss of a wind turbine and its implementation method. The system can automatically analyze the power loss of the wind turbine due to various reasons such as power grid power restrictions, unit failures, unit maintenance, and reduced performance during the operation of the wind turbine, and can automatically analyze the root cause of performance loss and give maintenance suggestions, thereby effectively improving the power generation of the unit. In the system for automatically identifying the root cause of power loss of a wind turbine, the unit is first divided by analyzing the unit status data, and then the various power losses of the unit are analyzed based on the theoretical power curve, wind speed and power generation data of the unit; then, by inputting the unit operation data, the root cause of performance loss is automatically analyzed and calculated, and then relevant operation and maintenance suggestions are given. This patent application does not take into account the correction of data, and there are still problems with the accuracy of the analysis. Summary of the invention

[0004] In order to overcome the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a method and device for automatically analyzing the causes of power loss of wind turbines. By calculating the power loss and extracting multi-dimensional features, and identifying the loss causes of multi-dimensional features through an integrated learning method, it is possible to process massive wind turbine operation data in real time, accurately identify the causes of power loss, and give optimization suggestions in a timely manner. It can significantly improve the operating efficiency of wind farms.

[0005] To achieve the above object, the technical solution adopted by the present invention is: In a first aspect, the present invention provides a method for automatically analyzing the cause of wind turbine power loss, comprising the following steps: Collect the operation data, environmental data and status data of the fan; Model the wind speed range in sections; Correction of air density and turbulence data; Calculate the power loss and extract multi-dimensional features, wherein the multi-dimensional features include wind condition features, power features, environmental features and motion features; Identifying loss causes of multi-dimensional features via ensemble learning.

[0006] Optionally, after collecting the data, outlier detection, data completion and standardization are performed on the collected data.

[0007] Optionally, an expert knowledge base information repository is established to select corresponding optimization suggestions and evaluate the implementation effect based on the identified loss causes.

[0008] Optionally, the correction coefficient K1 for air density correction is calculated as follows: ; The calculation formula of turbulence correction coefficient K2 is: ; where ρ is the air density, I is the standard turbulence intensity, It is an adaptive coefficient that takes into account the influence of wind speed distribution characteristics and terrain factors.

[0009] Optionally, the steps of the ensemble learning method include: constructing an integrated framework including a random forest classifier, a gradient boosting classifier and an XGBoost classifier; using each classifier to handle high-dimensional feature classification, nonlinear pattern recognition and sample imbalance problems respectively; fusing the results through an improved soft voting mechanism, and dynamically adjusting the voting weight according to the historical performance of the classifier.

[0010] Optionally, the ensemble learning method processes the classification of high-dimensional features through a random forest classification method, processes the classification of nonlinear features through a gradient boosting classification method, and optimizes the unbalanced samples through an XGBoost classification method.

[0011] In a second aspect, the present invention provides a system for automatically analyzing the causes of power loss of a wind turbine, comprising: Data acquisition module, used to collect the operation data, environmental data and status data of the fan; Classification modeling module, used to model the wind speed range in sections; The first calculation module is used for correcting air density and turbulence data; A second calculation module is used to calculate the power loss and extract multi-dimensional features, wherein the multi-dimensional features include wind condition features, power features, environmental features and motion features; An analysis module for identifying loss causes of multi-dimensional features through ensemble learning.

[0012] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for automatically analyzing the cause of power loss of a wind turbine when executing the computer program.

[0013] 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 method for automatically analyzing the cause of power loss of a wind turbine is implemented.

[0014] In a fifth aspect, the present invention provides a computer program product including a computer-readable medium, wherein the computer-readable medium contains a computer-readable program code, and the program code executes the method for automatically analyzing the cause of power loss of a wind turbine.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention effectively solves the problem of massive data processing by establishing an automated data collection and preprocessing mechanism, combined with an improved quartile rule and hierarchical completion strategy. The dynamic correction mechanism of intelligent segmented modeling and multi-factor coupling is adopted to significantly improve the fitting accuracy of the power curve. Through multi-dimensional feature extraction and integrated learning framework, the automatic identification of the cause of power loss is achieved, and the analysis accuracy rate reaches more than 95%. The optimization suggestions generated based on the expert knowledge base and multi-objective optimization algorithm have strong practical guiding significance, and the feasibility of the optimization suggestions exceeds 90%.

[0016] Furthermore, the system of the present invention adopts a distributed system architecture design, so that the system has strong scalability and fault tolerance, can support parallel analysis of hundreds of wind turbines at the same time, and the response time is controlled at the millisecond level.

[0017] Furthermore, the present invention realizes the full process automation of wind turbine power loss analysis, greatly improves the analysis efficiency, reduces operation and maintenance costs, and has important practical value for improving the overall operational efficiency of wind farms. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way.

[0019] In the attached picture: Figure 1 A schematic flow chart of a method for automatically analyzing the cause of power loss of a wind turbine according to Embodiment 1 of the present invention; Figure 2 Schematic diagram of the process of data collection and preprocessing steps of Example 1 of the present invention; Figure 3 It is a flowchart of the power curve modeling and correction steps of Embodiment 1 of the present invention; Figure 4 Schematic diagram of the flow of power loss analysis steps in Example 1 of the present invention; Figure 5 This is a schematic diagram of the system architecture of Example 2 of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the technical solutions in 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 described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0022] It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present application may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names. In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined. The present invention is described in detail below with reference to the accompanying drawings.

[0023] A method for automatically analyzing the cause of power loss of a wind turbine according to the present invention comprises the following steps: Collect the operation data, environmental data and status data of the fan; Model the wind speed range in sections; Correction of air density and turbulence data; Calculate the power loss and extract multi-dimensional features, wherein the multi-dimensional features include wind condition features, power features, environmental features and motion features; Identifying loss causes of multi-dimensional features via ensemble learning.

[0024] The present invention effectively solves the problem of massive data processing by establishing an automated data collection and preprocessing mechanism, combined with an improved quartile rule and hierarchical completion strategy. The dynamic correction mechanism of intelligent segmented modeling and multi-factor coupling is adopted to significantly improve the fitting accuracy of the power curve. Through multi-dimensional feature extraction and integrated learning framework, the automatic identification of the cause of power loss is achieved. The optimization suggestions generated based on the expert knowledge base and multi-objective optimization algorithm have strong practical guiding significance.

[0025] Example 1 See also Figure 1 As shown in the flowchart of the method, the method realizes intelligent identification of wind turbine power loss causes and generation of optimization suggestions through a systematic data processing and analysis process. The entire analysis process includes four main steps: data collection and preprocessing, power curve modeling and correction, power loss analysis, and generation of optimization suggestions. These steps are closely linked to form a complete analysis chain.

[0026] A method for automatically analyzing the cause of wind turbine power loss, comprising the following steps: 1) Data collection and preprocessing steps, collecting fan operation data, environmental data and status data, and performing outlier detection, data completion and standardization on the data; 2) Power curve modeling and correction steps: segmented modeling of the wind speed range, and air density correction and turbulence correction; 3) Power loss analysis step: calculate the power loss, extract multi-dimensional features, and identify the cause of the loss through ensemble learning method; 4) Optimization suggestion generation step: Generate optimization suggestions based on the identified loss causes and evaluate the implementation effect.

[0027] Furthermore, in the data collection and preprocessing steps: the wind turbine operation data includes wind speed, power output, wind direction, generator speed and pitch angle; the environmental data includes temperature, humidity and air pressure; the status data includes equipment vibration, component temperature, operation status code and fault alarm information.

[0028] Optionally, the sampling interval for data collection is 10 minutes.

[0029] like Figure 2 As shown, the system collects three types of key data at the same time: operation data, environmental data and status data. The present invention uses a 10-minute sampling interval for data collection. The selection of this time interval not only ensures the timeliness of the data, but also avoids the burden on the system caused by too frequent sampling.

[0030] In order to ensure the accuracy of subsequent analysis, the present invention designs a complete set of data preprocessing mechanisms. First, in the outlier detection link, the present invention adopts an improved quartile rule in a breakthrough way. This method not only calculates the first quartile Q1 and the third quartile Q3 of the data, but also dynamically adjusts the calculation method of the interquartile range IQR to make the determination of outliers more accurate. Specifically, the system automatically selects the appropriate determination boundary according to the characteristics of different data types through an adaptive threshold adjustment mechanism, thereby improving the accuracy of outlier identification.

[0031] Furthermore, the outlier detection adopts an improved quartile rule, specifically including: calculating the first quartile Q1 and the third quartile Q3 of the data; calculating the interquartile range IQR=Q3-Q1; dynamically adjusting the decision boundary according to the data type; marking the data less than Q1-1.5IQR or greater than Q3+1.5IQR as outliers.

[0032] In the data completion link, the present invention proposes a hierarchical completion strategy based on time scale. When the data missing time does not exceed 30 minutes, the system adopts an improved linear interpolation algorithm, which not only considers the continuity of the time series, but also introduces the weight factor of the data fluctuation characteristics, so that the interpolation result is more in line with the actual situation. For long-term data missing exceeding 30 minutes, the system adopts an intelligent historical data matching algorithm, which calculates the similarity of multi-dimensional features and selects the closest data segment from the historical data for supplementation, ensuring the rationality of the supplemented data.

[0033] In the power curve modeling and correction link, the present invention proposes an innovative segmented adaptive modeling method. Figure 3 As shown in the figure, this method first intelligently segments the wind speed range and automatically determines the optimal segmentation points through a mathematical model, overcoming the limitations of the traditional fixed segmentation method. In the curve fitting process, differentiated modeling strategies are adopted according to the characteristics of different wind speed segments: for the startup segment, a progressive fitting model considering the characteristics of the unit is adopted; for the climbing segment, a polynomial model corrected for the wind shear effect is introduced; for the rated segment, a dynamic reference value determination method based on measured data statistics is adopted. This segmented adaptive modeling method significantly improves the fitting accuracy of the power curve.

[0034] In the working condition correction link, the present invention proposes a dynamic correction mechanism of multi-factor coupling. First, in terms of air density correction, the present invention breaks through the traditional single correction coefficient method and establishes a dynamic coupling correction model based on temperature and air pressure.

[0035] pass Calculate the actual air density and introduce a nonlinear correction factor , where the exponent 0.33 is obtained through optimization of a large amount of experimental data. This correction method improves the accuracy by more than 15% compared with the traditional method.

[0036] In terms of turbulence correction, the present invention proposes an adaptive turbulence correction algorithm. This algorithm not only takes into account the standard turbulence intensity , and also introduces the influence of wind speed distribution characteristics and terrain factors. Correction coefficient By improving the calculation formula Get, where is an adaptive coefficient, which is dynamically adjusted according to the real-time working conditions. In this way, the system can more accurately reflect the impact of turbulence on fan performance.

[0037] Where ρ is the air density, T is the temperature, It is an adaptive coefficient that takes into account the influence of wind speed distribution characteristics and terrain factors.

[0038] Reference Figure 4 The power loss analysis process shown in the figure, the present invention designs a multi-dimensional collaborative analysis framework. In the power loss calculation module, the system innovatively adopts the time series cumulative comparison method, which not only calculates the instantaneous power difference, but also considers the dynamic characteristics of the power curve, significantly improving the accuracy of loss quantification. In particular, the system introduces a time weighting mechanism in the calculation process, differentiates the losses in different time periods, and more accurately reflects the actual impact of the losses.

[0039] Furthermore, the feature extraction in the power loss analysis step includes: wind condition characteristics: including wind speed, wind direction parameters, wind speed shear index and wind direction disturbance factor; power characteristics: including power output fluctuation rate and dynamic characteristics based on power curve deviation; environmental characteristics: including temperature gradient, air pressure change rate and atmospheric stability index; operation characteristics: including speed fluctuation, pitch angle change and equipment response characteristic index.

[0040] Furthermore, the ensemble learning method includes: constructing an integrated framework including a random forest classifier, a gradient boosting classifier and an XGBoost classifier; each classifier is responsible for processing high-dimensional feature classification, nonlinear pattern recognition and sample imbalance problems respectively; and the results are integrated through an improved soft voting mechanism, and the voting weight is dynamically adjusted according to the historical performance of the classifier.

[0041] Furthermore, the optimization suggestion generation step includes: screening potential optimization solutions from an expert knowledge base; comprehensively considering factors such as implementation cost, expected benefits, and technical feasibility through a multi-objective optimization algorithm; calculating the priority score of the optimization solution; and generating implementation suggestions including specific parameter adjustment values, maintenance plans, and expected improvement effects.

[0042] Furthermore, the method is implemented through a distributed system, which includes: a data acquisition unit: adopting a multi-threaded parallel acquisition mechanism; a preprocessing unit: adopting a streaming processing architecture; an analysis unit: adopting a microservice architecture, including a power curve modeling module and a loss analysis module; an optimization unit: adopting an intelligent decision-making system based on a rule engine; a data storage unit: adopting a distributed database architecture; and a communication unit: adopting a message queue mechanism.

[0043] In the feature extraction stage, the present invention introduces multiple deep features: in the wind condition feature, in addition to the basic wind speed and wind direction parameters, the wind speed shear index and wind direction disturbance factor are also calculated; in the power feature, the dynamic feature based on the power curve deviation is innovatively proposed; in the environmental feature, the atmospheric stability index is introduced; in the operation feature, the equipment response characteristic index is developed. The combination of these features provides comprehensive data support for the subsequent loss cause identification.

[0044] The loss cause identification module uses an innovative integrated learning framework. The framework contains three core classifiers: the random forest classifier is responsible for handling the classification problem of high-dimensional features, the gradient boosting classifier focuses on processing nonlinear patterns, and the XGBoost classifier is optimized for sample imbalance. The three classifiers are combined through an improved soft voting mechanism, and the voting weights are dynamically adjusted according to the historical performance of each classifier, which significantly improves the accuracy and stability of classification.

[0045] In the optimization suggestion generation link, the present invention proposes an intelligent decision support system. The system first selects potential optimization solutions from a pre-established expert knowledge base based on the category of loss causes. Then, through a multi-objective optimization algorithm, a priority score is assigned to each optimization solution by comprehensively considering factors such as implementation cost, expected benefits, and technical feasibility. Finally, the system will generate detailed implementation suggestions based on the actual situation of the wind farm, including specific parameter adjustment values, maintenance plans, and expected improvement effects.

[0046] Example 3 This embodiment provides a system for implementing the method of the above embodiment 1. Figure 5As shown in the figure, the system adopts a distributed architecture design and includes six functional units. The data acquisition unit is located at the upstream of the system and adopts an innovative multi-threaded parallel acquisition mechanism, which significantly improves the efficiency and reliability of data acquisition. The preprocessing unit adopts a streaming processing architecture and can process massive data in real time. The analysis unit is the core of the system and adopts a microservice architecture to decouple the power curve modeling module and the loss analysis module, thereby improving the flexibility and scalability of the system.

[0047] The optimization unit uses an intelligent decision-making system based on a rule engine, which can quickly generate optimization suggestions based on analysis results. The data storage unit uses a distributed database architecture, which not only ensures high data availability but also supports high concurrent access. The communication unit uses a message queue mechanism to ensure reliable communication and data synchronization between functional units.

[0048] The system architecture design of the present invention is significantly innovative: first, the distributed design improves the scalability and fault tolerance of the system; second, the microservice architecture allows each module of the system to be independently upgraded and optimized; finally, the streaming processing mechanism ensures the real-time performance of the system. According to actual tests, the system can support parallel analysis of hundreds of wind turbines at the same time, and the response time is controlled at the millisecond level.

[0049] The present invention realizes intelligent analysis and precise optimization of the causes of wind turbine power loss through the above technical scheme. Practical application shows that this method can improve the recognition accuracy of the causes of wind turbine power loss to more than 95%, and the feasibility of optimization suggestions reaches more than 90%, which has significant practical value for improving the power generation efficiency of wind farms. It should be noted that the above embodiments are only preferred implementation modes of the present invention, and the protection scope of the present invention is not limited thereto. Those skilled in the art may also make other modifications or improvements without departing from the technical scheme of the present invention, and these modifications or improvements should all fall within the protection scope of the present invention.

[0050] Example 4 The purpose of this embodiment is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for automatically analyzing the cause of power loss of a wind turbine when executing the computer program.

[0051] Example 5 The purpose of this embodiment is to provide 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 method for automatically analyzing the cause of power loss of a wind turbine is implemented.

[0052] Example 6 The purpose of this embodiment is to provide a computer program product including a computer-readable medium, on which a computer-readable program code is contained, and the program code executes the method for automatically analyzing the cause of power loss of a wind turbine.

[0053] The steps involved in the devices of the above embodiments 3, 4, 5 and 6 correspond to those of the method embodiment 1. For the specific implementation methods, please refer to the relevant description part of embodiment 1.

[0054] It should be understood by those skilled in the art that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt 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 codes. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0055] Unless otherwise specified, the working modes or control modes involved in the above embodiments are all conventional working modes or control modes in the art.

[0056] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application. Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for automatically analyzing the causes of wind turbine power loss, characterized in that: The following steps are involved: Collect the operation data, environmental data and status data of the fan; Model the wind speed range in sections; Correction of air density and turbulence data; Calculate the power loss and extract multi-dimensional features, wherein the multi-dimensional features include wind condition features, power features, environmental features and motion features; Identifying loss causes of multi-dimensional features via ensemble learning.

2. The method for automatically analyzing the causes of wind turbine power loss according to claim 1, characterized in that: After collecting the data, outlier detection, data completion and standardization are performed on the collected data.

3. The method for automatically analyzing the causes of wind turbine power loss according to claim 1, characterized in that: Establish an expert knowledge base information repository, select corresponding optimization suggestions and evaluate implementation effects based on the identified loss causes.

4. The method for automatically analyzing the causes of wind turbine power loss according to claim 1, characterized in that: The calculation formula of the correction coefficient K1 for air density correction is: ; The calculation formula of turbulence correction factor K2 is: ; where ρ is the air density, I is the standard turbulence intensity, It is an adaptive coefficient that takes into account the influence of wind speed distribution characteristics and terrain factors.

5. The method for automatically analyzing the causes of wind turbine power loss according to claim 1, characterized in that: The steps of the ensemble learning method include: constructing an integrated framework including a random forest classifier, a gradient boosting classifier and an XGBoost classifier; using each classifier to handle high-dimensional feature classification, nonlinear pattern recognition and sample imbalance problems respectively; fusing the results through an improved soft voting mechanism, and dynamically adjusting the voting weight according to the historical performance of the classifier.

6. The method for automatically analyzing the causes of wind turbine power loss according to claim 1, characterized in that: The ensemble learning method processes the classification of high-dimensional features through the random forest classification method, processes the classification of nonlinear features through the gradient boosting classification method, and optimizes unbalanced samples through the XGBoost classification method.

7. An automatic analysis system for the causes of wind turbine power loss, characterized in that: include: Data acquisition module, used to collect the operation data, environmental data and status data of the fan; Classification modeling module, used to model the wind speed range in sections; The first calculation module is used for correcting air density and turbulence data; A second calculation module is used to calculate the power loss and extract multi-dimensional features, wherein the multi-dimensional features include wind condition features, power features, environmental features and motion features; An analysis module for identifying loss causes of multi-dimensional features through ensemble learning.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for automatically analyzing the cause of power loss of a wind turbine as claimed in any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for automatically analyzing the cause of wind turbine power loss according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer readable medium, characterized in that The computer-readable medium contains computer-readable program code, and the program code executes the method for automatically analyzing the cause of power loss of a wind turbine as described in any one of claims 1-6.

Images