Wind generating set fault diagnosis method and system based on artificial intelligence

Abstract

The invention discloses a wind generating set fault diagnosis method and system based on artificial intelligence, and the method comprises the steps: obtaining and preprocessing the historical operation monitoring data of a wind generating set, extracting the features of the data, building an operation data prediction model based on the features and a preset neural network model, and carrying out the prediction, and obtaining operation prediction data; acquiring normal operation data of the wind generating set, and comparing and analyzing the normal operation data and the operation prediction data to determine difference parameters; evaluating the operation condition of the wind generating set based on the difference parameter to obtain an operation evaluation value, and diagnosing whether the wind generating set has a fault based on the operation evaluation value; and after the fault is diagnosed, the fault type of the wind generating set is diagnosed according to the difference parameters to generate corresponding early warning information. Fault diagnosis of the wind generating set is carried out by combining artificial intelligence prediction, intelligent and accurate diagnosis can be realized, the fault diagnosis efficiency is improved, and the reliability and safety of operation are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis of wind turbine generators, and particularly to a fault diagnosis method and system for wind turbine generators based on artificial intelligence. Background Art

[0002] With the rapid development of renewable energy, wind power generation, as one of the clean energies, has been widely used globally. However, faults in wind turbine generators have seriously affected their normal operation and power generation efficiency, resulting in production losses and increased maintenance costs. Therefore, developing an efficient and accurate fault diagnosis method is crucial for improving the reliability and operation efficiency of wind turbine generators.

[0003] However, traditional fault monitoring and warning methods usually rely on the experience and knowledge of professional maintenance personnel to diagnose faults. This method is easily affected by subjective factors, has high requirements for operators, is inefficient and costly, limiting the accuracy and reliability of diagnosis. Moreover, the amount of data generated by wind turbine generators is huge, and traditional methods often struggle to effectively process large-scale data, leading to low efficiency in information extraction and analysis, limiting the accuracy and timeliness of fault diagnosis. Additionally, traditional methods mainly diagnose faults that have already occurred and lack the ability to predict possible future faults, unable to take preventive measures to avoid fault occurrence in advance. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a fault diagnosis method and system for wind turbine generators based on artificial intelligence, including: Obtain the historical operation monitoring data of the wind turbine generator and preprocess the historical operation monitoring data; Extract the features of the preprocessed historical operation monitoring data, and construct an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; Obtain the normal operation data of the wind turbine generator, and compare and analyze the normal operation data with the operation prediction data to determine the difference parameters; Evaluate the operation condition of the wind turbine generator based on the difference parameters to obtain an operation evaluation value, and diagnose whether the wind turbine generator has a fault based on the operation evaluation value; After diagnosing that the wind turbine generator has a fault, diagnose the fault type of the wind turbine generator according to the difference parameters, and generate corresponding warning information according to the fault type.

[0005] Further, the obtaining the historical operation monitoring data of the wind turbine generator and preprocessing the historical operation monitoring data includes: Obtain the historical operation monitoring data of the wind turbine from the database, and preprocess the historical operation monitoring data. The preprocessing includes removing outliers and noise from the data, filling in missing values in the data, and normalizing the data.

[0006] Furthermore, extract the features of the preprocessed historical operation monitoring data, and construct an operation data prediction model based on the features and a preset neural network model to perform prediction, obtaining operation prediction data, including: Divide the preprocessed historical operation monitoring data into several groups of operation parameter data according to the parameter type, extract the features in each group of operation parameter data, and construct sub-datasets based on each group of operation parameter data and the corresponding features respectively; Input the sub-datasets into the corresponding preset sub-neural network models respectively to construct sub-operation prediction initial models corresponding to each group of operation parameter data; Divide the sub-datasets into a training set and a test set according to a preset ratio, and input the training set and the test set into the corresponding sub-operation prediction initial models respectively; Train and test the sub-operation prediction initial models respectively until the sub-operation prediction initial models meet the preset convergence conditions to obtain sub-operation prediction models corresponding to each group of operation parameter data; Fuse the sub-operation prediction models corresponding to each group of operation parameter data to obtain an operation data prediction model for the wind turbine; Obtain the current real-time operation monitoring data of the wind turbine, and input the real-time operation monitoring data into the operation data prediction model of the wind turbine to perform prediction, obtaining the operation prediction data of the wind turbine.

[0007] Furthermore, obtain the normal operation data of the wind turbine, and compare and analyze the normal operation data with the operation prediction data to determine the difference parameters, including: Obtain the normal operation data of the wind turbine from the database, and determine the same parameters in the normal operation data and the operation prediction data; Construct several parameter comparison curves based on the same parameters respectively, and take the curve corresponding to the normal operation data in the parameter comparison curves as the first curve, and the curve corresponding to the operation prediction data as the second curve; Compare the determined first curve and the second curve in each parameter comparison curve, determine the parameter comparison curves in which the first curve and the second curve are significantly different among several parameter comparison curves, and determine the parameters corresponding to these parameter comparison curves as the difference parameters.

[0008] Further, the comparison of the first curve and the second curve in each parameter comparison curve graph, and the determination of the parameter comparison curve graphs with obvious differences between the first curve and the second curve in several parameter comparison curve graphs include: Determine the non-overlapping curves of the first curve and the second curve in each parameter comparison curve graph, and determine the proportion of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves; Determine the parameter comparison curve graphs in which both the proportion of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves are greater than the corresponding preset thresholds as the parameter comparison curve graphs with obvious differences between the first curve and the second curve.

[0009] Further, the evaluation of the operation condition of the wind turbine based on the difference parameters to obtain an operation evaluation value includes: Determine the first curve and the second curve in the parameter comparison curve graph corresponding to the difference parameter, and calculate the variance difference and the average value difference between the first curve and the second curve; Evaluate and obtain the first evaluation value and the second evaluation value for the variance difference and the average value difference respectively, and add the first evaluation value and the second evaluation value to calculate the comprehensive evaluation value of the difference parameter; Determine the preset weight of the difference parameter, and calculate based on the preset weight and the comprehensive evaluation value of the difference parameter to obtain the operation evaluation value of the wind turbine. The calculation formula of the operation evaluation value of the wind turbine is: , where D is the operation evaluation value of the wind turbine, ai is the preset weight of the i-th difference parameter, Pi is the comprehensive evaluation value of the i-th difference parameter, and n is the number of difference parameters.

[0010] Further, the diagnosis of whether the wind turbine has a fault based on the operation evaluation value includes: Preset a fault threshold, and diagnose the relationship between the operation evaluation value and the fault threshold to determine whether the wind turbine has a fault; If the operation evaluation value exceeds the fault threshold, it is determined that the wind turbine has a fault; If the operation evaluation value does not exceed the fault threshold, it is determined that the wind turbine has no fault.

[0011] Further, after diagnosing that the wind turbine has a fault, diagnosing the fault type of the wind turbine according to the difference parameters includes: After diagnosing that the wind turbine has a fault, determine each difference parameter, and match each difference parameter with the fault parameter corresponding to the preset fault type; Diagnose the fault type of the wind turbine generator set by matching the difference parameter with the fault parameter.

[0012] Furthermore, generating corresponding warning information according to the fault type includes: Preset the correspondence between warning information and fault type. For each fault type, there is a corresponding warning information associated with it; Obtain the fault type of the wind turbine generator set, and based on the mapping relationship of the fault type to which the fault type belongs in the correspondence between warning information and fault type, select the warning information corresponding to the fault type and determine it as the warning information of the wind turbine generator set.

[0013] The present invention also provides an artificial intelligence-based fault diagnosis system for wind turbine generator sets, including: A preprocessing module for obtaining the historical operation monitoring data of the wind turbine generator set and preprocessing the historical operation monitoring data; A prediction module for extracting the features of the preprocessed historical operation monitoring data and constructing an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; A comparison module for obtaining the normal operation data of the wind turbine generator set and comparing and analyzing the normal operation data with the operation prediction data to determine the difference parameter; An evaluation module for evaluating the operation condition of the wind turbine generator set based on the difference parameter to obtain an operation evaluation value, and diagnosing whether the wind turbine generator set has a fault based on the operation evaluation value; A warning module for diagnosing the fault type of the wind turbine generator set according to the difference parameter after diagnosing that the wind turbine generator set has a fault, and generating corresponding warning information according to the fault type.

[0014] Compared with the prior art, the artificial intelligence-based fault diagnosis method and system for wind turbine generator sets in the embodiments of the present invention have the following beneficial effects: Through the preprocessing and feature extraction of historical operation data, and the operation data prediction combined with the neural network model, the present invention can realize the prediction and warning of possible faults of the wind turbine generator set, which helps to take measures in advance and reduce the impact of sudden faults on the operation of the generator set; By comparing and analyzing the difference parameter and evaluating the operation condition, the present invention can realize the real-time monitoring and diagnosis of the operation state of the wind turbine generator set, timely discover abnormal situations, and accurately diagnose the fault type, which helps to quickly respond and solve problems; Through the prediction and warning system, the present invention can reduce the unplanned downtime, avoid losses and improve the power generation efficiency. At the same time, timely diagnosing faults and generating warning information helps to reduce the troubleshooting time of maintenance personnel and reduce the maintenance cost; Through continuous monitoring, prediction, and diagnosis, the present invention can improve the reliability and stability of wind turbines, promptly detect and handle potential problems, contribute to ensuring the long-term stable operation of the system, and reduce the impact of faults on power generation capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic flow structure diagram of a fault diagnosis method for a wind turbine based on artificial intelligence in an embodiment of the present invention; Figure 2 is a schematic composition diagram of a fault diagnosis system for a wind turbine based on artificial intelligence in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following combines the drawings and embodiments to further describe in detail the specific implementation manners of the present application. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0017] In the description of the present application, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the platform or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present application.

[0018] The terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0019] In the description of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0020] Such as Figure 1As shown in the figure, in the embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided, including: S100: Obtain the historical operation monitoring data of the wind turbine and preprocess the historical operation monitoring data; S200: Extract the features of the preprocessed historical operation monitoring data, and construct an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; S300: Obtain the normal operation data of the wind turbine, and compare and analyze the normal operation data with the operation prediction data to determine the difference parameters; S400: Evaluate the operation condition of the wind turbine based on the difference parameters to obtain an operation evaluation value, and diagnose whether the wind turbine has a fault based on the operation evaluation value; S500: After diagnosing that the wind turbine has a fault, diagnose the fault type of the wind turbine according to the difference parameters, and generate a corresponding warning message according to the fault type.

[0021] Furthermore, through the preprocessing and feature extraction of historical operation data, combined with the operation data prediction of the neural network model, the present invention can realize the prediction and early warning of possible faults of the wind turbine, which helps to take measures in advance to reduce the impact of sudden faults on the operation of the generator set; by comparing and analyzing the difference parameters and evaluating the operation condition, the present invention can realize the real-time monitoring and diagnosis of the operation state of the wind turbine, timely discover abnormal situations, and accurately diagnose the fault type, which helps to quickly respond and solve problems; through the prediction and early warning system, the present invention can reduce the unplanned downtime, avoid losses and improve the power generation efficiency. At the same time, timely diagnosing faults and generating warning messages helps to reduce the troubleshooting time of maintenance personnel and reduce the maintenance cost; through continuous monitoring, prediction and diagnosis, the present invention can improve the reliability and stability of the wind turbine, timely discover and handle potential problems, which helps to ensure the long-term stable operation of the system and reduce the impact of faults on the power generation capacity.

[0022] In the embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. The obtaining of the historical operation monitoring data of the wind turbine and the preprocessing of the historical operation monitoring data include: obtaining the historical operation monitoring data of the wind turbine from a database and preprocessing the historical operation monitoring data. The preprocessing includes removing outliers and noise from the data, filling in missing values in the data, and normalizing the data.

[0023] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. The method includes extracting features of preprocessed historical operation monitoring data, and constructing an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data, including: dividing the preprocessed historical operation monitoring data into several groups of operation parameter data groups according to parameter types, extracting features in each group of operation parameter data, and respectively constructing sub-datasets based on each group of operation parameter data and the corresponding features; inputting the sub-datasets into the corresponding preset sub-neural network models respectively to construct initial sub-operation prediction models corresponding to each group of operation parameter data; dividing the sub-datasets into a training set and a test set according to a preset ratio, and inputting the training set and the test set into the corresponding initial sub-operation prediction models respectively; training and testing the initial sub-operation prediction models respectively until the initial sub-operation prediction models meet the preset convergence condition to obtain sub-operation prediction models corresponding to each group of operation parameter data; fusing the sub-operation prediction models corresponding to each group of operation parameter data to obtain an operation data prediction model for the wind turbine; obtaining the current real-time operation monitoring data of the wind turbine, and inputting the real-time operation monitoring data into the operation data prediction model of the wind turbine for prediction to obtain the operation prediction data of the wind turbine.

[0024] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. The method includes obtaining the normal operation data of the wind turbine, and comparing and analyzing the normal operation data with the operation prediction data to determine differential parameters, including: obtaining the normal operation data of the wind turbine from a database, and determining the same parameters in the normal operation data and the operation prediction data; respectively constructing several parameter comparison curves based on the same parameters, taking the curve corresponding to the normal operation data in the parameter comparison curve as the first curve, and taking the curve corresponding to the operation prediction data as the second curve; comparing the first curve and the second curve determined in each parameter comparison curve, determining the parameter comparison curves in which the first curve and the second curve are significantly different in several parameter comparison curves, and determining the parameters corresponding to these parameter comparison curves as differential parameters.

[0025] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. Comparing the first curve and the second curve determined in each parameter comparison curve graph, and determining the parameter comparison curve graphs in which the first curve and the second curve are significantly different among a plurality of parameter comparison curve graphs, including: determining the non-overlapping curves of the first curve and the second curve in each parameter comparison curve graph, and determining the occupancy ratio of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves; determining the parameter comparison curve graphs in which both the occupancy ratio of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves are greater than the corresponding preset thresholds as the parameter comparison curve graphs in which the first curve and the second curve are significantly different.

[0026] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. Evaluating the operating condition of the wind turbine based on the difference parameters to obtain an operation evaluation value, including: determining the first curve and the second curve in the parameter comparison curve graph corresponding to the difference parameters, and calculating the variance difference and the average difference between the first curve and the second curve; respectively evaluating and obtaining a first evaluation value and a second evaluation value for the variance difference and the average difference, and adding the first evaluation value and the second evaluation value to calculate the comprehensive evaluation value of the difference parameters; determining the preset weight of the difference parameters, and calculating based on the preset weight and the comprehensive evaluation value of the difference parameters to obtain the operation evaluation value of the wind turbine. The calculation formula for the operation evaluation value of the wind turbine is: , where D is the operation evaluation value of the wind turbine, ai is the preset weight of the i-th difference parameter, Pi is the comprehensive evaluation value of the i-th difference parameter, and n is the number of difference parameters.

[0027] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. Diagnosing whether the wind turbine has a fault based on the operation evaluation value, including: presetting a fault threshold, and diagnosing the relationship between the operation evaluation value and the fault threshold to determine whether the wind turbine has a fault; if the operation evaluation value exceeds the fault threshold, it is determined that the wind turbine has a fault; if the operation evaluation value does not exceed the fault threshold, it is determined that the wind turbine has no fault.

[0028] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. After diagnosing that the wind turbine has a fault, diagnosing the fault type of the wind turbine according to the difference parameters, including: after diagnosing that the wind turbine has a fault, determining each difference parameter, and matching each difference parameter with the fault parameters corresponding to the preset fault types; diagnosing the fault type in which the difference parameter and the fault parameter match as the fault type of the wind turbine.

[0029] In an embodiment of the present application, a fault diagnosis method for a wind turbine based on artificial intelligence is provided. Generating corresponding warning information according to the fault type includes: presetting the correspondence between warning information and fault type. For each fault type, corresponding warning information is associated; obtaining the fault type of the wind turbine, and based on the mapping relationship of the fault type to which the fault type belongs in the correspondence between warning information and fault type, selecting the warning information corresponding to the fault type and determining it as the warning information of the wind turbine.

[0030] As Figure 2 shown, in an embodiment of the present application, a fault diagnosis system for a wind turbine based on artificial intelligence is provided, including: a preprocessing module for obtaining historical operation monitoring data of the wind turbine and preprocessing the historical operation monitoring data; a prediction module for extracting features of the preprocessed historical operation monitoring data and constructing an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; a comparison module for obtaining normal operation data of the wind turbine and comparing and analyzing the normal operation data with the operation prediction data to determine a difference parameter; an evaluation module for evaluating the operation condition of the wind turbine based on the difference parameter to obtain an operation evaluation value and diagnosing whether the wind turbine has a fault based on the operation evaluation value; a warning module for diagnosing the fault type of the wind turbine according to the difference parameter after diagnosing that the wind turbine has a fault, and generating corresponding warning information according to the fault type.

[0031] In summary, the embodiments of the present invention provide a fault diagnosis method and system for a wind turbine based on artificial intelligence, which include: obtaining and preprocessing the historical operation monitoring data of the wind turbine, extracting its features, and constructing an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; obtaining the normal operation data of the wind turbine and comparing and analyzing the normal operation data with the operation prediction data to determine a difference parameter; evaluating the operation condition of the wind turbine based on the difference parameter to obtain an operation evaluation value and diagnosing whether the wind turbine has a fault based on it; after diagnosing a fault, diagnosing the fault type of the wind turbine according to the difference parameter to generate corresponding warning information. The present invention can achieve intelligent and precise diagnosis by combining artificial intelligence prediction for fault diagnosis of wind turbines, improve the efficiency of fault diagnosis, and ensure the reliability and safety of operation.

[0032] Finally, it should be noted that: Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

[0033] The above is only one embodiment of the present invention, but it should not be used to limit the scope of the present invention. Any structural changes made in accordance with the present invention, as long as the essence of the present invention is not lost, should be regarded as falling within the protection scope of the present invention and being restricted. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related descriptions of the above-described platform can refer to the corresponding process in the foregoing platform embodiment and will not be elaborated herein.

[0034] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, so that a process, platform, article, or device / platform comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or also includes elements inherent in these processes, platforms, articles, or devices / platforms.

[0035] So far, the technical solutions of the present invention have been described in conjunction with the further embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to closely related technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0036] The above is only the preferred embodiment of the present invention and is not used to limit the protection scope of the present invention.

Claims

1. A fault diagnosis method for wind turbine generators based on artificial intelligence, characterized in that, Including: Obtain the historical operation monitoring data of the wind turbine generator set, and preprocess the historical operation monitoring data; Extract the features of the preprocessed historical operation monitoring data, and construct an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; Obtain the normal operation data of the wind turbine generator set, and compare and analyze the normal operation data with the operation prediction data to determine the difference parameters; Evaluate the operation condition of the wind turbine generator set based on the difference parameters to obtain an operation evaluation value, and diagnose whether the wind turbine generator set has a fault based on the operation evaluation value; After diagnosing that the wind turbine generator set has a fault, diagnose the fault type of the wind turbine generator set according to the difference parameters, and generate a corresponding warning message according to the fault type.

2. The fault diagnosis method of a wind turbine based on artificial intelligence according to claim 1, wherein The obtaining the historical operation monitoring data of the wind turbine generator set and preprocessing the historical operation monitoring data includes: Obtain the historical operation monitoring data of the wind turbine generator set from the database, and preprocess the historical operation monitoring data. The preprocessing includes removing outliers and noise in the data, filling in missing values in the data, and normalizing the data.

3. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 2, characterized in that The extracting the features of the preprocessed historical operation monitoring data and constructing an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data includes: Divide the preprocessed historical operation monitoring data into several groups of operation parameter data groups according to the parameter type, extract the features in each group of operation parameter data, and construct sub-datasets based on each group of operation parameter data and the corresponding features respectively; Input the sub-datasets into the corresponding preset sub-neural network models respectively to construct sub-operation prediction initial models corresponding to each group of operation parameter data; Divide the sub-datasets into a training set and a test set according to a preset ratio, and input the training set and the test set into the corresponding sub-operation prediction initial models respectively; Train and test the sub-operation prediction initial models respectively until the sub-operation prediction initial models meet the preset convergence conditions to obtain sub-operation prediction models corresponding to each group of operation parameter data; Fuse the sub-operation prediction models corresponding to each group of operation parameter data to obtain an operation data prediction model of the wind turbine generator set; Obtain the current real-time operation monitoring data of the wind turbine generator set, and input the real-time operation monitoring data into the operation data prediction model of the wind turbine generator set for prediction to obtain the operation prediction data of the wind turbine generator set.

4. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 3, characterized in that, The obtaining the normal operation data of the wind turbine generator set and comparing and analyzing the normal operation data with the operation prediction data to determine the difference parameters includes: Obtain the normal operation data of the wind turbine generator set from the database, and determine the same parameters in the normal operation data and the operation prediction data; Construct several parameter comparison curve graphs based on the same parameters respectively, and use the curve corresponding to the normal operation data in the parameter comparison curve graph as the first curve, and use the curve corresponding to the operation prediction data as the second curve; Compare the first curve and the second curve determined in each parameter comparison curve graph, determine the parameter comparison curve graphs in which the first curve and the second curve are significantly different among several parameter comparison curve graphs, and determine the parameters corresponding to these parameter comparison curve graphs as the differential parameters.

5. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 4, characterized in that, The comparison of the first curve and the second curve determined in each parameter comparison curve graph, and the determination of the parameter comparison curve graphs in which the first curve and the second curve are significantly different among several parameter comparison curve graphs include: Determine the non-overlapping curves of the first curve and the second curve in each parameter comparison curve graph, and determine the proportion of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves; Determine the parameter comparison curve graphs in which both the proportion of the non-overlapping curves and the total area enclosed by the first curve and the second curve in the non-overlapping curves are greater than the corresponding preset thresholds as the parameter comparison curve graphs in which the first curve and the second curve are significantly different.

6. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 4, characterized in that The evaluation of the operating condition of the wind turbine based on the differential parameters to obtain the operating evaluation value includes: Determine the first curve and the second curve in the parameter comparison curve graph corresponding to the differential parameter, and calculate the variance difference and the average difference between the first curve and the second curve; Evaluate and obtain the first evaluation value and the second evaluation value for the variance difference and the average difference respectively, and add the first evaluation value and the second evaluation value to calculate the comprehensive evaluation value of the differential parameter; Determine the preset weight of the differential parameter, and calculate based on the preset weight and the comprehensive evaluation value of the differential parameter to obtain the operating evaluation value of the wind turbine. The calculation formula for the operating evaluation value of the wind turbine is: ， where D is the operating evaluation value of the wind turbine, ai is the preset weight of the i-th differential parameter, Pi is the comprehensive evaluation value of the i-th differential parameter, and n is the number of differential parameters.

7. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 6, characterized in that The diagnosis of whether the wind turbine has a fault based on the operating evaluation value includes: Preset a fault threshold, and diagnose the relationship between the operating evaluation value and the fault threshold to determine whether the wind turbine has a fault; If the operating evaluation value exceeds the fault threshold, it is determined that the wind turbine has a fault; If the operating evaluation value does not exceed the fault threshold, it is determined that the wind turbine has no fault.

8. A fault diagnosis method for a wind turbine generator set based on artificial intelligence according to claim 7, characterized in that, After diagnosing that the wind turbine has a fault, diagnose the fault type of the wind turbine according to the differential parameters, including: After diagnosing that the wind turbine has a fault, determine each differential parameter, and match each differential parameter with the fault parameter corresponding to the preset fault type; Diagnose the fault type in which the differential parameter and the fault parameter match as the fault type of the wind turbine.

9. A fault diagnosis method for a wind turbine generator based on artificial intelligence according to claim 8, characterized in that, The generation of the corresponding warning information according to the fault type includes: Preset the warning information - fault type correspondence. For each fault type in the warning information - fault type correspondence, a corresponding warning information is associated; Obtain the fault type of the wind turbine, and select the warning information corresponding to the fault type as the warning information of the wind turbine based on the mapping relationship of the fault type to which the fault type belongs in the warning information - fault type correspondence.

10. A fault diagnosis system for a wind turbine generator set based on artificial intelligence, characterized in that, including: A preprocessing module, which is used to obtain the historical operation monitoring data of the wind turbine generator set and preprocess the historical operation monitoring data; A prediction module, which is used to extract the features of the preprocessed historical operation monitoring data, and build an operation data prediction model based on the features and a preset neural network model for prediction to obtain operation prediction data; A comparison module, which is used to obtain the normal operation data of the wind turbine generator set, and compare and analyze the normal operation data with the operation prediction data to determine the difference parameters; An evaluation module, which is used to evaluate the operation condition of the wind turbine generator set based on the difference parameters to obtain an operation evaluation value, and diagnose whether the wind turbine generator set has a fault based on the operation evaluation value; An early warning module, which is used to diagnose the fault type of the wind turbine generator set according to the difference parameters after diagnosing that the wind turbine generator set has a fault, and generate corresponding early warning information according to the fault type.

Images