Wind power generation abnormal data analysis method and system

Abstract

The invention discloses a wind power generation abnormal data analysis method and system, and the method comprises the steps: obtaining historical wind power generation data of a wind power station, and carrying out the preprocessing of the historical wind power generation data; analyzing the preprocessed historical wind power generation data, and determining abnormal parameter data with anomalies in the historical wind power generation data; extracting data features of the abnormal parameter data, and determining abnormal type data according to the data features; establishing an exception classification model based on exception type data corresponding to the exception parameter data and the data features, and determining a current exception type according to the exception classification model; determining an exception evaluation index of the exception type, and evaluating the exception type through the exception evaluation index to obtain an exception degree value; and determining the abnormal level of the wind power station according to the abnormal degree value. According to the scheme, the abnormity severity degree can be accurately and comprehensively analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a method and system for analyzing abnormal data of wind power generation. Background Art

[0002] With the development and application of renewable energy, wind power generation, as a form of clean energy, has become an important energy source worldwide. However, various abnormal situations may occur during the operation of wind power generation equipment, including equipment failure, performance degradation, changes in meteorological conditions, etc. Therefore, analyzing and monitoring abnormal wind power data can help improve the reliability, safety and operating efficiency of wind power equipment.

[0003] However, traditional abnormal data analysis methods usually rely on manual experience and rules to identify abnormal situations, which are easily affected by human subjective factors, resulting in low processing efficiency and poor accuracy. In addition, traditional methods only rely on simple statistical analysis to identify abnormal situations and cannot comprehensively analyze and understand abnormal patterns in the data. At the same time, traditional methods find it difficult to conduct a detailed and comprehensive assessment of abnormal situations and it is difficult to determine abnormal situations in wind power generation. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method and system for analyzing abnormal data of wind power generation.

[0005] A first aspect of the present invention provides a method for analyzing abnormal data of wind power generation, comprising:

[0006] Obtain the historical wind power generation data of the wind power station and pre-process the historical wind power generation data;

[0007] Analyze the pre-processed historical wind power generation data to determine abnormal parameter data that are abnormal in the historical wind power generation data;

[0008] Extracting data features of the abnormal parameter data, and determining abnormal type data of the abnormal parameter data according to the data features;

[0009] An abnormality classification model of the wind power station is established based on the abnormality type data and data features corresponding to the abnormal parameter data, and the current abnormality type of the wind power station is determined according to the abnormality classification model;

[0010] Determine an abnormality evaluation index of the abnormality type, and evaluate the abnormality degree of the abnormality type by using the abnormality evaluation index to obtain an abnormality degree value of the abnormality type;

[0011] The abnormality level of the wind power station is determined according to the abnormality degree value.

[0012] Furthermore, the preprocessing of the historical wind power generation data includes:

[0013] Clean the historical wind power data to delete or correct the wrong values, missing values ​​and duplicate values ​​in the data;

[0014] The cleaned historical wind power generation data are smoothed and normalized.

[0015] Furthermore, the analyzing the pre-processed historical wind power generation data to determine abnormal parameter data having abnormalities in the historical wind power generation data includes:

[0016] The pre-processed historical wind power generation data is divided into multiple parameter data sets according to the monitoring parameter type, and the average value of the data in each parameter data set is calculated respectively;

[0017] The difference between each data in each parameter data set and the data average is calculated, and the data with a difference greater than a preset threshold is screened out as abnormal parameter data.

[0018] Further, extracting data features of the abnormal parameter data and determining abnormal type data of the abnormal parameter data according to the data features includes:

[0019] Analyze the difference between abnormal parameter data and normal data in historical wind power generation data, and extract data features from the abnormal parameter data based on the difference;

[0020] The extracted data features are clustered based on the k-means clustering algorithm, the number of cluster centers is determined according to the type of abnormal parameter data, and different abnormal types in the abnormal parameter data are identified and determined according to the clustering results.

[0021] Furthermore, the abnormal type data and data features corresponding to the abnormal parameter data are used to establish an abnormal classification model for the wind power station, and the current abnormal type of the wind power station is determined according to the abnormal classification model, including:

[0022] The abnormal type corresponding to the abnormal parameter data is used as a label, the data feature is used as an input feature, and a data set is established based on the data corresponding to the label and the input feature;

[0023] Build an abnormal classification model for wind power plants based on the data set and the preset neural network model;

[0024] The data set is divided into a training set and a test set according to a certain ratio, and the training set and the test set are input into the anomaly classification model to train and test the anomaly classification model until the anomaly classification model meets the preset convergence conditions;

[0025] The current real-time wind power generation data of the wind power station is obtained, and the real-time wind power generation data is input into the trained anomaly classification model, and the anomaly classification model determines the anomaly type corresponding to the real-time wind power generation data.

[0026] Further, the abnormality evaluation index of the abnormality type is determined, and the abnormality degree of the abnormality type is evaluated by the abnormality evaluation index to obtain the abnormality degree value of the abnormality type, including:

[0027] Set several anomaly assessment indicators for different anomaly types respectively, analyze the importance of the anomaly assessment indicators, and determine the weight coefficient of the anomaly type according to the importance of the anomaly assessment indicators;

[0028] According to a number of anomaly assessment indicators, the corresponding anomaly types are evaluated and valued respectively to obtain an anomaly assessment value of the anomaly type;

[0029] The weight coefficient and the abnormal evaluation value of the abnormal type are calculated comprehensively by the weighted average method to obtain the abnormal degree value of the abnormal type. The expression is:

[0030]

[0031] Among them, S is the abnormality degree value of the abnormal type, αi is the weight coefficient of the i-th abnormal type, and Xi is the abnormal evaluation value of the i-th abnormal type.

[0032] Furthermore, analyzing the importance of the abnormality assessment index and determining the weight coefficient of the abnormality type according to the importance of the abnormality assessment index include:

[0033] The expert assignment method is used to assign and score the anomaly assessment index, and the scores of multiple experts are obtained. The average value of all scores is calculated to obtain the importance of the anomaly assessment index, and the weight coefficient of the anomaly type is determined according to the importance of the anomaly assessment index. The expression is:

[0034]

[0035] Among them, α is the weight coefficient of the abnormal type, t c is the importance of the abnormal evaluation index, t s is the average importance of all abnormal evaluation indicators, and exp represents the natural exponential function.

[0036] Further, determining the abnormality level of the wind power station according to the abnormality degree value includes:

[0037] A plurality of abnormality degree interval-abnormality level mapping tables are preset, and a plurality of abnormality degree interval-abnormality level corresponding relationships are associated with a plurality of abnormality levels for each abnormality degree interval of the wind power station;

[0038] Obtaining an abnormality level value of the wind power station, and determining an abnormality level of the wind power station based on a mapping relationship between an abnormality level interval in which the abnormality level value is located in a preset abnormality level interval-abnormality level mapping table;

[0039] If the abnormality level value is in the first abnormality level interval, the wind power station is at the first abnormality level;

[0040] If the abnormality level value is in the second abnormality level interval, the wind power station is at the second abnormality level;

[0041] If the abnormality level value is in the third abnormality level interval, the wind power station is at the third abnormality level.

[0042] The second aspect of the present invention further provides a wind power generation abnormal data analysis system, comprising:

[0043] An acquisition module is used to acquire the past historical wind power generation data of the wind power station and pre-process the historical wind power generation data;

[0044] An analysis module is used to analyze the pre-processed historical wind power generation data to determine abnormal parameter data that are abnormal in the historical wind power generation data;

[0045] A processing module, used for extracting data features of abnormal parameter data, and determining abnormal type data of the abnormal parameter data according to the data features;

[0046] A prediction module, used to establish an abnormal classification model of the wind power station based on the abnormal type data and data features corresponding to the abnormal parameter data, and determine the current abnormal type of the wind power station according to the abnormal classification model;

[0047] An evaluation module is used to determine an abnormality evaluation index of an abnormality type, and evaluate the abnormality degree of the abnormality type through the abnormality evaluation index to obtain an abnormality degree value of the abnormality type;

[0048] The determination module is used to determine the abnormality level of the wind power station according to the abnormality level value.

[0049] Compared with the prior art, the wind power generation abnormal data analysis method and system provided by the present invention have the following beneficial effects:

[0050] The present invention determines the current abnormality type by establishing an abnormality classification model, calculates the abnormality degree value by the abnormality evaluation index, and determines the abnormality level of the wind power station according to the abnormality degree value, so as to accurately and comprehensively analyze the severity of the abnormality. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 1 is a schematic diagram of the process structure of the abnormal data analysis method for wind power generation according to an embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of the composition of the wind power generation abnormal data analysis system in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0054] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply 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 understood as a limitation on the present application.

[0055] The terms "second" 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. Therefore, a feature defined with "second" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "multiple" means two or more.

[0056] In the description of this 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 can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0057] like Figure 1As shown, in an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, including: S100: acquiring the past historical wind power generation data of a wind power station, and preprocessing the historical wind power generation data; S200: analyzing the preprocessed historical wind power generation data, and determining abnormal parameter data having abnormalities in the historical wind power generation data; S300: extracting data features of the abnormal parameter data, and determining abnormal type data of the abnormal parameter data according to the data features; S400: establishing an abnormal classification model of the wind power station based on the abnormal type data and data features corresponding to the abnormal parameter data, and determining the current abnormal type of the wind power station according to the abnormal classification model; S500: determining an abnormal evaluation index of the abnormal type, and evaluating the abnormal degree of the abnormal type through the abnormal evaluation index, and obtaining an abnormal degree value of the abnormal type; S600: determining the abnormal level of the wind power station according to the abnormal degree value.

[0058] Furthermore, the present invention can automatically identify abnormal situations in historical wind power generation data through machine learning and data analysis technology, thereby reducing the manual identification and processing of abnormal data; the present invention can divide abnormal data into different abnormal types by constructing an abnormal classification model, thereby helping engineers to better understand the types and characteristics of abnormal situations; the present invention can monitor and analyze real-time data of wind power stations in combination with real-time data processing technology, and promptly discover and respond to abnormal situations; the present invention can objectively evaluate the severity of abnormal situations by determining abnormal evaluation indicators and calculating abnormal degree values, thereby providing a basis for subsequent processing and decision-making; the present invention determines the abnormal level of the wind power station according to the abnormal degree value, thereby helping engineers and managers to quickly understand the severity of the abnormal situation and take corresponding measures.

[0059] In an embodiment of the present application, a method for analyzing abnormal wind power generation data is provided, wherein the preprocessing of historical wind power generation data includes: cleaning the historical wind power generation data, deleting or correcting erroneous values, missing values ​​and duplicate values ​​in the data; and smoothing and normalizing the cleaned historical wind power generation data.

[0060] Specifically, the historical wind power generation data is cleaned, and the values ​​that exceed the reasonable range are detected and processed, such as wind speed exceeding the limit value, power output being negative, etc. In combination with domain knowledge and actual conditions, the abnormal values ​​are corrected or marked, and missing values ​​are detected and processed by interpolation, filling the mean / median, using adjacent values, etc. For severely missing data, it is necessary to consider whether to discard the corresponding records or perform other processing according to the actual situation, detect and process duplicate data records, and ensure the uniqueness and accuracy of the data; the cleaned historical wind power generation data is smoothed and normalized. For time series data, moving average, exponential weighted moving average and other methods can be used for smoothing to reduce noise interference and highlight data trends. The data is scaled to the range of [0,1] and normalized by subtracting the minimum value and dividing by the range (maximum value minus minimum value). Through data cleaning, smoothing and normalization, the quality and availability of historical wind power generation data can be improved, laying a good foundation for subsequent analysis and modeling.

[0061] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein the preprocessed historical wind power generation data is analyzed to determine abnormal parameter data that is abnormal in the historical wind power generation data, including: dividing the preprocessed historical wind power generation data into multiple parameter data sets according to the monitoring parameter type, and calculating the data average value in each parameter data set respectively; calculating the difference between each data in each parameter data set and the data average value, and filtering out the data whose difference is greater than a preset threshold as abnormal parameter data.

[0062] Specifically, the historical wind power generation data is divided into multiple data sets according to the monitoring parameter type. For example, if the data includes parameters such as wind speed and power output, the wind speed data and the power output data can be divided into two data sets respectively. Then, for each parameter data set, the average value of the data is calculated; next, for each parameter data set, the difference between each data and the average value is calculated, for example, for each data point, the difference between its value and the average value of the parameter data set is calculated; then, a preset threshold is set, and according to the set threshold, data with a difference greater than the preset threshold is screened out, and these data are abnormal parameter data.

[0063] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein data features of abnormal parameter data are extracted, and abnormal type data of the abnormal parameter data is determined based on the data features, including: analyzing the difference between abnormal parameter data and normal data in historical wind power generation data, and extracting data features from the abnormal parameter data based on the difference; clustering the extracted data features based on a k-means clustering algorithm, determining the number of clustering centers based on the type of abnormal parameter data, and identifying and determining different abnormal types in the abnormal parameter data based on the clustering results.

[0064] Specifically, a comparative analysis is conducted on the abnormal parameter data and the normal data, and the differences in the distribution of the abnormal parameter data and the normal data on the monitoring parameters are found out by comparing the data distribution characteristics; based on the analysis results, the abnormal parameter data features that need to be extracted are determined, and the extracted abnormal parameter data features are clustered using the k-means clustering algorithm. Before clustering, the number of clustering centers is determined, and for the type identification of the abnormal parameter data, the elbow method (E l bowMethod) is used to determine the optimal number of clustering centers; according to the clustering results, different abnormal types in the abnormal parameter data are identified and determined.

[0065] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein an abnormal classification model of a wind power station is established based on abnormal type data and data features corresponding to abnormal parameter data, and the current abnormal type of the wind power station is determined according to the abnormal classification model, including: taking the abnormal type corresponding to the abnormal parameter data as a label, taking the data features as input features, and establishing a data set according to the data corresponding to the label and the input features; constructing an abnormal classification model of the wind power station based on the data set and a preset neural network model; dividing the data set into a training set and a test set according to a certain ratio, and inputting the training set and the test set into the abnormal classification model, training and testing the abnormal classification model until the abnormal classification model meets the preset convergence condition; acquiring the current real-time wind power generation data of the wind power station, and inputting the real-time wind power generation data into the trained abnormal classification model, and determining the abnormal type corresponding to the real-time wind power generation data by the abnormal classification model.

[0066] Specifically, according to the abnormal type corresponding to the abnormal parameter data, the data features are used as input features and the abnormal type is used as a label to establish a data set, and each data includes the input feature and the corresponding abnormal type label; using the preset neural network model, the input feature is used as the input layer, and the output is the corresponding abnormal type label; the constructed data set is divided into a training set and a test set according to a certain ratio, with the ratio of 70% training set and 30% test set, and the training set is input into the abnormal classification model for training until the model meets the preset convergence conditions; the current real-time wind power generation data of the wind power station is obtained, the real-time data is used as input, and reasoning is performed through the trained abnormal classification model to determine the abnormal type corresponding to the real-time wind power generation data.

[0067] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein the abnormal evaluation index of the abnormal type is determined, and the abnormal degree of the abnormal type is evaluated by the abnormal evaluation index to obtain the abnormal degree value of the abnormal type, including: setting a number of abnormal evaluation indexes of different abnormal types, analyzing the importance of the abnormal evaluation index, and determining the weight coefficient of the abnormal type according to the importance of the abnormal evaluation index; evaluating and valuing the corresponding abnormal type according to the number of abnormal evaluation indicators to obtain the abnormal evaluation value of the abnormal type; and comprehensively calculating the weight coefficient and the abnormal evaluation value of the abnormal type by the weighted average method to obtain the abnormal degree value of the abnormal type, and the expression is:

[0068]

[0069] Among them, S is the abnormality degree value of the abnormal type, αi is the weight coefficient of the i-th abnormal type, and Xi is the abnormal evaluation value of the i-th abnormal type.

[0070] Specifically, for different anomaly types, several anomaly evaluation indicators are set; according to the importance of the anomaly evaluation indicators, the weight coefficient of the anomaly type is determined; according to the set anomaly evaluation indicators, the corresponding anomaly type is evaluated and valued to obtain an anomaly evaluation value of each anomaly type; through the weighted average method, the weight coefficient and the anomaly evaluation value of the anomaly type are comprehensively calculated to obtain the anomaly degree value of the anomaly type. Specifically, the anomaly evaluation value of each anomaly type is multiplied by the corresponding weight coefficient, and then the results are added to obtain the anomaly degree value of the anomaly type.

[0071] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein the importance of the abnormal evaluation index is analyzed, and the weight coefficient of the abnormal type is determined according to the importance of the abnormal evaluation index, including: assigning and scoring the abnormal evaluation index by an expert assignment method to obtain the scores of multiple experts, calculating the average of all scores to obtain the importance of the abnormal evaluation index, and determining the weight coefficient of the abnormal type according to the importance of the abnormal evaluation index, and the expression is:

[0072]

[0073] Among them, α is the weight coefficient of the abnormal type, t c is the importance of the abnormal evaluation index, t s is the average importance of all abnormal evaluation indicators, and exp represents the natural exponential function.

[0074] Specifically, multiple domain experts are invited to assign values ​​and scores to the anomaly assessment indicators. Each expert can score each anomaly assessment indicator based on his or her professional knowledge and experience, usually on a scale of 1 to 10, where 10 points means very important and 1 point means not important. After collecting the scores of multiple experts, the scores of each indicator are averaged to obtain the average importance score of each indicator. The purpose of this is to comprehensively consider the opinions of multiple experts and obtain a more objective indicator importance. The weight coefficient of the anomaly type is determined based on the calculated average importance score of the anomaly assessment indicator.

[0075] In an embodiment of the present application, a method for analyzing abnormal data of wind power generation is provided, wherein the abnormal level of a wind power station is determined according to an abnormality level value, including: presetting a plurality of abnormality level interval-abnormality level mapping tables, wherein a plurality of abnormality level interval-abnormality level correspondences are associated with a plurality of abnormality levels for each abnormality level interval of the wind power station; obtaining the abnormality level value of the wind power station, and determining the abnormality level of the wind power station based on a mapping relationship between the abnormality level interval in which the abnormality level value is located in the preset abnormality level interval-abnormality level mapping table; if the abnormality level value is in a first abnormality level interval, the wind power station is at a first abnormality level; if the abnormality level value is in a second abnormality level interval, the wind power station is at a second abnormality level; and if the abnormality level value is in a third abnormality level interval, the wind power station is at a third abnormality level.

[0076] Specifically, a mapping table of abnormality degree interval-abnormality level is preset, and the table contains abnormality degree intervals and corresponding abnormality levels. For example, the abnormality degree interval can be a numerical range, and the abnormality level can be a corresponding text identifier, such as "level one abnormality level", "level two abnormality level", and "level three abnormality level". For each abnormal situation of the wind power station, its abnormality degree value is obtained, and the abnormality level of the wind power station is determined based on the mapping relationship between the abnormality degree interval where the abnormality degree value is located in the preset abnormality degree interval-abnormality level mapping table. Specifically, the abnormality degree value is compared with the preset abnormality degree interval to find the corresponding abnormality level.

[0077] like Figure 2As shown, in an embodiment of the present application, a wind power generation abnormal data analysis system is provided, including: an acquisition module, used to acquire the past historical wind power generation data of a wind power station and preprocess the historical wind power generation data; an analysis module, used to analyze the preprocessed historical wind power generation data, and determine abnormal parameter data with abnormalities in the historical wind power generation data; a processing module, used to extract data features of the abnormal parameter data, and determine abnormal type data of the abnormal parameter data according to the data features; a prediction module, used to establish an abnormal classification model of the wind power station based on the abnormal type data and data features corresponding to the abnormal parameter data, and determine the current abnormal type of the wind power station according to the abnormal classification model; an evaluation module, used to determine an abnormal evaluation index of the abnormal type, and evaluate the abnormal degree of the abnormal type through the abnormal evaluation index to obtain an abnormal degree value of the abnormal type; a determination module, used to determine the abnormal level of the wind power station according to the abnormal degree value.

[0078] In summary, an embodiment of the present invention provides a method and system for analyzing abnormal data of wind power generation, which includes: obtaining historical wind power generation data of a wind power station and preprocessing it; analyzing the preprocessed historical wind power generation data to determine abnormal parameter data in which there are abnormalities; extracting data features of the abnormal parameter data, and determining the abnormal type according to the data features; establishing an abnormal classification model based on the abnormal type data and data features corresponding to the abnormal parameter data, and determining the current abnormal type according to the abnormal classification model; determining an abnormal evaluation index of the abnormal type, and evaluating the abnormal type by the abnormal evaluation index to obtain an abnormal degree value; determining the abnormal level of the wind power station according to the abnormal degree value. The present invention determines the current abnormal type by establishing an abnormal classification model, calculates the abnormal degree value by the abnormal evaluation index, and determines the abnormal level of the wind power station according to the abnormal degree value, so as to accurately and comprehensively analyze the severity of the abnormality.

[0079] Finally, it should be noted that it is obvious that those skilled in the art can make various changes and modifications 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 equivalents, the present invention is also intended to include these modifications and variations.

[0080] The above is only an example of implementation of the present invention, but it cannot be used to limit the scope of the present invention. Any structural changes made according to the present invention, as long as they do not lose the essence of the present invention, should be regarded as falling within the scope of protection of the present invention and being restricted. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process and related instructions of the platform described above can refer to the corresponding process in the aforementioned platform embodiment, and will not be repeated here.

[0081] The term "comprises" or any other similar term is intended to cover a non-exclusive inclusion such that a process, platform, article, or apparatus / platform that includes a list of elements includes not only those elements but also other elements not expressly listed or inherent to such process, platform, article, or apparatus / platform.

[0082] So far, the technical solutions of the present invention have been described in conjunction with the further embodiments shown in the accompanying drawings, but 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 the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0083] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention.

Claims

1. A method for analyzing abnormal data of wind power generation, characterized in that: include: Obtain the historical wind power generation data of the wind power station and pre-process the historical wind power generation data; Analyze the pre-processed historical wind power generation data to determine abnormal parameter data that are abnormal in the historical wind power generation data; Extracting data features of the abnormal parameter data, and determining abnormal type data of the abnormal parameter data according to the data features; An abnormality classification model of the wind power station is established based on the abnormality type data and data features corresponding to the abnormal parameter data, and the current abnormality type of the wind power station is determined according to the abnormality classification model; Determine an abnormality evaluation index of the abnormality type, and evaluate the abnormality degree of the abnormality type by using the abnormality evaluation index to obtain an abnormality degree value of the abnormality type; The abnormality level of the wind power station is determined according to the abnormality degree value.

2. A method for analyzing abnormal data of wind power generation according to claim 1, characterized in that: The preprocessing of the historical wind power generation data includes: Clean the historical wind power data to delete or correct the wrong values, missing values ​​and duplicate values ​​in the data; The cleaned historical wind power generation data are smoothed and normalized.

3. A method for analyzing abnormal data of wind power generation according to claim 2, characterized in that: The analyzing the pre-processed historical wind power generation data to determine abnormal parameter data existing in the historical wind power generation data includes: The pre-processed historical wind power generation data is divided into multiple parameter data sets according to the monitoring parameter type, and the average value of the data in each parameter data set is calculated respectively; The difference between each data in each parameter data set and the data average is calculated, and the data with a difference greater than a preset threshold is screened out as abnormal parameter data.

4. A method for analyzing abnormal data of wind power generation according to claim 3, characterized in that: The step of extracting data features of the abnormal parameter data and determining abnormal type data of the abnormal parameter data according to the data features includes: Analyze the difference between abnormal parameter data and normal data in historical wind power generation data, and extract data features from the abnormal parameter data based on the difference; The extracted data features are clustered based on the k-means clustering algorithm, the number of cluster centers is determined according to the type of abnormal parameter data, and different abnormal types in the abnormal parameter data are identified and determined according to the clustering results.

5. The method for analyzing abnormal data of wind power generation according to claim 1, characterized in that: The abnormality classification model of the wind power station is established based on the abnormality type data and data features corresponding to the abnormal parameter data, and the current abnormality type of the wind power station is determined according to the abnormality classification model, including: The abnormal type corresponding to the abnormal parameter data is used as a label, the data feature is used as an input feature, and a data set is established based on the data corresponding to the label and the input feature; Build an abnormal classification model for wind power plants based on the data set and the preset neural network model; The data set is divided into a training set and a test set according to a certain ratio, and the training set and the test set are input into the anomaly classification model to train and test the anomaly classification model until the anomaly classification model meets the preset convergence conditions; The current real-time wind power generation data of the wind power station is obtained, and the real-time wind power generation data is input into the trained anomaly classification model, and the anomaly classification model determines the anomaly type corresponding to the real-time wind power generation data.

6. A method for analyzing abnormal data of wind power generation according to claim 5, characterized in that: The step of determining an abnormality evaluation index of an abnormality type and evaluating the abnormality degree of the abnormality type by using the abnormality evaluation index to obtain an abnormality degree value of the abnormality type includes: Set several anomaly assessment indicators for different anomaly types respectively, analyze the importance of the anomaly assessment indicators, and determine the weight coefficient of the anomaly type according to the importance of the anomaly assessment indicators; According to a number of anomaly assessment indicators, the corresponding anomaly types are evaluated and valued respectively to obtain an anomaly assessment value of the anomaly type; The weight coefficient and the abnormal evaluation value of the abnormal type are calculated comprehensively by the weighted average method to obtain the abnormal degree value of the abnormal type. The expression is: Among them, S is the abnormality degree value of the abnormal type, αi is the weight coefficient of the i-th abnormal type, and Xi is the abnormal evaluation value of the i-th abnormal type.

7. A method for analyzing abnormal data of wind power generation according to claim 6, characterized in that: The analyzing the importance of the abnormality assessment index and determining the weight coefficient of the abnormality type according to the importance of the abnormality assessment index include: The expert assignment method is used to assign and score the anomaly assessment index, and the scores of multiple experts are obtained. The average value of all scores is calculated to obtain the importance of the anomaly assessment index, and the weight coefficient of the anomaly type is determined according to the importance of the anomaly assessment index. The expression is: Among them, α is the weight coefficient of the abnormal type, t c is the importance of the abnormal evaluation index, t s is the average importance of all abnormal evaluation indicators, and exp represents the natural exponential function.

8. A method for analyzing abnormal data of wind power generation according to claim 6, characterized in that: Determining the abnormality level of the wind power station according to the abnormality degree value includes: A plurality of abnormality degree interval-abnormality level mapping tables are preset, and a plurality of abnormality degree interval-abnormality level corresponding relationships are associated with a plurality of abnormality levels for each abnormality degree interval of the wind power station; Acquire the abnormality level value of the wind power station, and determine the abnormality level of the wind power station based on the mapping relationship between the abnormality level interval in which the abnormality level value is located in a preset abnormality level interval-abnormality level mapping table; If the abnormality level value is in the first abnormality level interval, the wind power station is at the first abnormality level; If the abnormality level value is in the second abnormality level interval, the wind power station is at the second abnormality level; If the abnormality level value is in the third abnormality level interval, the wind power station is at the third abnormality level.

9. A wind power generation abnormal data analysis system, characterized in that: include: An acquisition module is used to acquire the past historical wind power generation data of the wind power station and pre-process the historical wind power generation data; An analysis module is used to analyze the pre-processed historical wind power generation data to determine abnormal parameter data that are abnormal in the historical wind power generation data; A processing module, used for extracting data features of abnormal parameter data, and determining abnormal type data of the abnormal parameter data according to the data features; A prediction module, used to establish an abnormal classification model of the wind power station based on the abnormal type data and data features corresponding to the abnormal parameter data, and determine the current abnormal type of the wind power station according to the abnormal classification model; An evaluation module is used to determine an abnormality evaluation index of an abnormality type, and evaluate the abnormality degree of the abnormality type through the abnormality evaluation index to obtain an abnormality degree value of the abnormality type; The determination module is used to determine the abnormality level of the wind power station according to the abnormality level value.

Images