A fan abnormal data identification method, system, device and medium under complex working conditions

Abstract

The application provides a fan abnormal data identification method, system, device and medium under complex working conditions, comprising the following steps: obtaining fan account data and fan operation data; filtering the fan account data and the fan operation data based on the fan operation principle, removing the shutdown state data to obtain a first filtered data set; processing the first filtered data based on a constant value abnormal data identification method of continuous difference calculation, removing the constant value abnormal data to obtain a second filtered data set; processing the second filtered data based on a limited power abnormal data identification method, removing the limited power abnormal data to obtain a third filtered data set; processing the third filtered data set based on a grid distribution abnormal data identification method, removing the peripheral scattered abnormal data to obtain a fourth filtered data set, and completing fan abnormal data identification.

Description

Technical Field

[0001] This invention belongs to the field of wind power operation detection technology, specifically relating to a method, system, equipment, and medium for identifying abnormal wind turbine data under complex operating conditions. Background Technology

[0002] As a relatively mature green new energy source, wind power has achieved rapid growth in technological innovation and installed capacity. However, this growth has also brought new challenges to wind farm operation and maintenance management, economic assessment, and safe production. Due to the complex and harsh environment in which wind turbines operate and the constraints imposed by power transmission and communication networks, the data collected by SCADA systems contains a large number of constant values ​​caused by sensor anomalies or communication interruptions, as well as a large amount of accumulated anomaly data due to power limitations. High-quality wind power data is crucial for wind turbine performance evaluation, condition monitoring, power prediction, and health diagnosis. Furthermore, variations in geographical location, manufacturer, turbine model, and commissioning time result in significant individual differences among wind turbines, placing high demands on the generalization ability of anomaly handling logic.

[0003] Existing related technologies mainly employ three approaches: statistical analysis, traditional machine learning, and image recognition. Among these, statistical analysis methods are ineffective for large numbers of clustered outlier data points; traditional machine learning methods face significant challenges in generalization and model parameter determination; and image recognition methods suffer from the inability to distinguish false anomalies near wind speed and low processing efficiency. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a method, system, device, and medium for identifying abnormal wind turbine data under complex operating conditions, thereby solving the technical problem that existing related technologies cannot accurately and efficiently identify abnormal wind turbine operation data.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for identifying abnormal data from wind turbines under complex operating conditions includes the following steps:

[0007] Obtain wind turbine ledger data and wind turbine operation data;

[0008] Based on the wind turbine operating principle, the wind turbine ledger data and wind turbine operating data are filtered to remove shutdown status data, resulting in a first-filtered dataset.

[0009] The constant value anomaly data identification method based on continuous difference calculation processes the first filtered data to remove constant value anomalies and obtain a second filtered dataset.

[0010] Based on the identification method for power limit anomaly data, the secondary filtered data is processed to remove the power limit anomaly data, resulting in a tertiary filtered dataset.

[0011] The grid-based abnormal data identification method processes the three-stage filtered dataset to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data.

[0012] Furthermore, the wind turbine ledger data includes the wind turbine cut-in wind speed w. in Cut-out wind speed w out Rated wind speed w rated and rated power p rated ;

[0013] The wind turbine operating data consists of aggregated average data of the wind turbine at 10-minute intervals over three months, including: time, wind speed, generator active power, pitch angle #1, pitch angle #2, and pitch angle #3.

[0014] Furthermore, the process of filtering the wind turbine ledger data and wind turbine operation data based on the wind turbine operating principle to remove shutdown status data and obtain a filtered dataset is as follows:

[0015] Set the shutdown status data filtering conditions, which include:

[0016] Wind speed value at cut-in wind speed w in -1 to cut-out wind speed w out Outside the range;

[0017] The average active power of the generator is 1.1 times the rated power p. rated Outside the range;

[0018] The average value of pitch angle 1, pitch angle 2 and pitch angle 3 is greater than 25.

[0019] Furthermore, the constant value outlier identification method based on continuous difference calculation processes the primary filtered data to remove constant value outliers and obtain the secondary filtered dataset as follows:

[0020] Sort the dataset according to the chronological order of time T in a single filtered dataset;

[0021] The second, third, and fourth order backward difference values ​​of the average wind speed W are solved sequentially to obtain the results of the wind speed difference values ​​at each order.

[0022] The 2nd, 3rd, and 4th order backward difference values ​​of the generator active power P are solved sequentially to obtain the results of the generator active power difference values ​​at each order.

[0023] If there exists a data point where the 2nd, 3rd, and 4th order backward difference values ​​of wind speed or generator active power are all zero, then the current data is identified as a constant value outlier, and the corresponding current data identified as a constant value outlier is deleted, resulting in a secondary filtered dataset.

[0024] Furthermore, the method for identifying power-limited anomaly data involves processing the secondary filtered data to remove the power-limited anomaly data, resulting in a tertiary filtered dataset.

[0025] In the secondary filtering dataset, the wind speed is selected based on the cut-in wind speed w. in and cut-in wind speed w in Data set with a +1 interval;

[0026] Obtain the pitch angles 1#, 2#, and 3# from the filtered dataset and form a pitch angle set. Filter the pitch angle set again to select data in the interval between the 20th and 80th percentiles, and calculate the mean and standard deviation of the pitch angles in this interval.

[0027] If the active power of a generator in a certain data point is less than 0.9 times the rated power, and the average value of any pitch angle 1, pitch angle 2, and pitch angle 3 is greater than the sum of the average pitch angle and the standard deviation, then the current data is identified as power-limited abnormal data, and the current data corresponding to the data identified as power-limited abnormal data is deleted, resulting in a secondary filtered dataset.

[0028] Furthermore, the formula for the method of identifying power-limited abnormal data is expressed as follows:

[0029]

[0030] Where D3 is the dataset with three-stage filtering, D2 is the dataset with two-stage filtering, and p rated For rated power, a mean The average pitch angle, a std D2(a) represents the standard deviation. n,i In the equation, n represents the nth propeller of the motor, and i represents the time.

[0031] Furthermore, the grid-based anomaly data identification method processes the three-stage filtered dataset to remove scattered anomaly data from the surrounding areas, resulting in a four-stage filtered dataset as follows:

[0032] Set the average wind speed as the x-axis and the average generator active power as the y-axis, and divide the average wind speed range [w] by 0.25 intervals. in -1, w out Divide the generator into equal intervals; and with intervals of 100, define the average active power range of the generator within the range [0, 0.9*p]. ratedThe intervals are divided equally to form a wind speed-generator active power grid;

[0033] Each data point in the three filtered datasets is mapped to the wind speed-generator active power grid, and the frequency of data contained in each grid is calculated respectively;

[0034] Filter out all data in the grid whose frequency is less than or equal to 5;

[0035] Select the average active power range of each generator in sequence, and in each range, obtain the corresponding grid frequency, and calculate the frequency of the generator in the active power range;

[0036] The frequencies are sorted in descending order, and the frequencies corresponding to the first n grids and the grids with frequencies greater than 0.9 are selected. The remaining grid data are then discarded.

[0037] The average wind speed range for each generator is selected sequentially, and the corresponding grid frequency is obtained in each range. The frequency of the generator in the average wind speed range is then calculated.

[0038] The frequencies are sorted in descending order, and the frequencies corresponding to the first n grids and the grids with frequencies greater than 0.9 are selected. The remaining grid data are removed; the four-stage filtered dataset is obtained, and the abnormal data of the wind turbine is identified.

[0039] A system for identifying abnormal data from wind turbines under complex operating conditions includes:

[0040] The acquisition unit is configured as follows:

[0041] Used to obtain wind turbine ledger data and wind turbine operation data;

[0042] The primary filter unit is configured as follows:

[0043] This is used to filter the wind turbine ledger data and wind turbine operation data based on the wind turbine operating principle, remove shutdown status data, and obtain a first-filtered dataset.

[0044] The secondary filtration unit is configured as follows:

[0045] A method for identifying constant value anomaly data based on continuous difference calculation processes the primary filtered data to remove constant value anomaly data and obtain a secondary filtered dataset.

[0046] The three-stage filtering unit is configured as follows:

[0047] The method for identifying power-limited anomaly data processes the secondary filtered data to remove the power-limited anomaly data, resulting in a tertiary filtered dataset.

[0048] The four-stage filtering and output unit is configured as follows:

[0049] The method for identifying abnormal data based on grid distribution processes the three-stage filtered dataset to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data.

[0050] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for identifying abnormal wind turbine data under complex operating conditions.

[0051] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for identifying abnormal wind turbine data under complex operating conditions.

[0052] Compared with the prior art, the present invention has the following beneficial technical effects:

[0053] This invention provides a method, system, device, and medium for identifying abnormal wind turbine data under complex operating conditions, comprising the following steps: acquiring wind turbine ledger data and wind turbine operation data; filtering the wind turbine ledger data and wind turbine operation data based on the wind turbine operation principle to remove shutdown status data, obtaining a primary filtered dataset; processing the primary filtered data based on a constant value abnormal data identification method using continuous difference calculation to remove constant value abnormal data, obtaining a secondary filtered dataset; processing the secondary filtered data based on a power-limited abnormal data identification method to remove power-limited abnormal data, obtaining a tertiary filtered dataset; processing the tertiary filtered dataset based on a grid distribution abnormal data identification method to remove peripheral scattered abnormal data, obtaining a quaternary filtered dataset, thus completing the identification of abnormal wind turbine data. This application proposes a constant value abnormal data identification method and a power-limited abnormal data identification method based on the root causes of abnormal data, wind turbine mechanisms, and data representation characteristics, respectively. These methods can effectively identify the above two types of abnormal data. Furthermore, by mining the distribution characteristics of grid-based data and comprehensively applying frequency and relative frequency representations, the identification and processing of abnormal data are rapidly achieved, enabling accurate and efficient identification of abnormal wind turbine operation data. Attached Figure Description

[0054] Figure 1 A flowchart of a method for identifying abnormal data of a wind turbine under complex operating conditions, according to an embodiment of this disclosure, is shown.

[0055] Figure 2 A schematic diagram of the structure of a wind turbine abnormal data identification system under complex operating conditions according to an embodiment of the present disclosure is shown. Detailed Implementation

[0056] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0057] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0058] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0059] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0060] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0061] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0062] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0063] Figure 1 A flowchart of a method 100 for identifying abnormal wind turbine data under complex operating conditions, according to an embodiment of this disclosure, is shown. Figure 1 As shown, it includes the following steps:

[0064] Step S101: Obtain wind turbine ledger data and wind turbine operation data;

[0065] Preferably, in this embodiment of the disclosure, the wind turbine ledger data includes the wind turbine cut-in wind speed w. in Cut-out wind speed w out Rated wind speed w rated and rated power o rated ;

[0066] The wind turbine operating data consists of aggregated average data of the wind turbine at 10-minute intervals over three months, including: time, wind speed, generator active power, pitch angle #1, pitch angle #2, and pitch angle #3.

[0067] Specifically, the time is represented as T = [t1, t2, ..., t n The wind speed, generator active power, pitch angle 1, pitch angle 2, and pitch angle 3 constitute a data set D, where D = [T, W, P, A1, A2, A3]. The five data acquisition items include:

[0068] Wind speed W = [w1, w2, ..., w n ];

[0069] Generator active power P = [p1, p2, ..., p n ];

[0070] #1 Pitch Angle A1=[a 1,1 ,a 1,2 ,...,a 1,n 】;

[0071] #2 propeller pitch angle A2=[a 2,1 ,a 2,2 ,…,a 2,n ];

[0072] And the pitch angle A3 of propeller #3 = [a 3,1 ,a 3,2 ,…,a 3,n ], where n represents the number of data entries in the dataset.

[0073] Step S102: Based on the wind turbine operating principle, filter the wind turbine ledger data and wind turbine operating data to remove shutdown status data and obtain a first-filtered dataset.

[0074] Preferably, in this embodiment of the disclosure, the process of obtaining a filtered dataset is as follows:

[0075] Set the shutdown status data filtering conditions, which include:

[0076] Wind speed value at cut-in wind speed win -1 to cut-out wind speed w out Outside the range;

[0077] The average active power of the generator is 1.1 times the rated power p. rated Outside the range;

[0078] The average value of pitch angle 1, pitch angle 2 and pitch angle 3 is greater than 25.

[0079] In other words, the filtering conditions for shutdown status data are divided into:

[0080]

[0081] In the formula, D1 represents the dataset after the shutdown status data has been processed; D(w i D(p) represents the wind speed at any time i in dataset D; i D(a) represents the active power of the generator at any time i in the dataset D; 1,i ) represents the propeller pitch angle at any time i in dataset D; D(a 2,i () represents the propeller pitch angle at any time i in dataset D; D(a) 3,i ) represents the propeller pitch angle at any time i in dataset D.

[0082] Step S103: Based on the constant value anomaly data identification method of continuous difference calculation, the primary filtered data is processed to remove constant value anomaly data and obtain the secondary filtered dataset.

[0083] Preferably, in this embodiment of the disclosure, the constant value anomaly data identification method based on continuous difference calculation processes the primary filtered data to remove constant value anomalies and obtain the secondary filtered dataset as follows:

[0084] Sort the dataset according to the chronological order of time T in a single filtered dataset;

[0085] The second, third, and fourth order backward difference values ​​are successively calculated for the average wind speed W to obtain the results of each order of wind speed difference; the specific calculation formula is as follows:

[0086] ΔD1(w i,k ) = w i -w i-k ;

[0087] In the above formula, k represents the order; i represents the i-th data point; w i Indicates the i-th wind speed; w i-k Indicates the wind speed of the i-th wind; ΔD1(w i,k ) represents the k-th order backward difference value of the wind speed at the i-th data point in dataset D1.

[0088] The 2nd, 3rd, and 4th order backward difference values ​​of the generator's active power P are solved sequentially to obtain the results of each order difference value of the generator's active power; the specific calculation formulas are as follows:

[0089] ΔD1(p i,k ) = p i -p i-k

[0090] In the above formula, p i p represents the i-th wind speed; i-k Indicates the wind speed of the i-th wind; ΔD1(p i,k ) represents the k-th order backward difference value of the generator active power at the i-th data point in dataset D1.

[0091] If there exists a data point where the 2nd, 3rd, and 4th order backward difference values ​​of wind speed or generator active power are all zero, then the current data is identified as a constant value outlier, and the corresponding current data identified as a constant value outlier is deleted, resulting in a secondary filtered dataset.

[0092] In other words, the calculation formula for the secondary filtered dataset can be expressed as:

[0093]

[0094] Step S104: Based on the method for identifying power-limited abnormal data, the secondary filtered data is processed to remove the power-limited abnormal data, resulting in a tertiary filtered dataset.

[0095] Preferably, in this embodiment of the disclosure, the method for identifying power-limited anomaly data involves processing the secondary filtered data to remove the power-limited anomaly data and obtaining the tertiary filtered dataset as follows:

[0096] In the secondary filtering dataset, the wind speed is selected based on the cut-in wind speed w. in and cut-in wind speed w in The dataset for the +1 interval is shown in the following formula:

[0097] D 2,1 ∈D2(w i )≥w in ∩D2(w i )≤w in +1

[0098] In the formula, D 2,1 This represents the data set that meets the above filtering criteria.

[0099] Obtain the pitch angles 1#, 2#, and 3# from the filtered dataset and form a pitch angle set. Filter the pitch angle set again to select data in the interval between the 20th and 80th percentiles, and calculate the mean and standard deviation of the pitch angles in this interval.

[0100] If the active power of a generator in a certain data point is less than 0.9 times the rated power, and the average value of any pitch angle 1, pitch angle 2, and pitch angle 3 is greater than the sum of the average pitch angle and the standard deviation, then the current data is identified as power-limited abnormal data, and the current data corresponding to the data identified as power-limited abnormal data is deleted, resulting in a three-stage filtered dataset.

[0101] Specifically, the formula for the method of identifying power-limited abnormal data is expressed as follows:

[0102]

[0103] Where D3 is the dataset with three-stage filtering, D2 is the dataset with two-stage filtering, and p rated For rated power, a mean The average pitch angle, a std D2(a) represents the standard deviation. n,i In the equation, n represents the nth propeller of the motor, and i represents the time.

[0104] Step S105: Based on the grid distribution abnormal data identification method, the three-stage filtered dataset is processed to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data.

[0105] Preferably, in this embodiment of the disclosure, the anomaly data identification method based on grid distribution processes the three-stage filtered dataset to remove scattered anomaly data from the surrounding areas, resulting in a four-stage filtered dataset as follows:

[0106] Set the average wind speed as the x-axis and the average generator active power as the y-axis, and divide the average wind speed range [w] by 0.25 intervals. in -1, w out Divide the generator into equal intervals; and with intervals of 100, define the average active power range of the generator within the range [0, 0.9*p]. rated The intervals are divided equally to form a wind speed-generator active power grid;

[0107] Each data point in the three filtered datasets is mapped to the wind speed-generator active power grid, and the frequency of data contained in each grid is calculated respectively;

[0108] Filter out all data in the grid whose frequency is less than or equal to 5;

[0109] Select the average active power range of each generator in sequence, and in each range, obtain the corresponding grid frequency, and calculate the frequency of the generator in the active power range;

[0110] The frequencies are sorted in descending order, and the frequencies corresponding to the first n grids and the grids with frequencies greater than 0.9 are selected. The remaining grid data are then discarded.

[0111] The average wind speed range for each generator is selected sequentially, and the corresponding grid frequency is obtained in each range. The frequency of the generator in the average wind speed range is then calculated.

[0112] The frequencies are sorted in descending order, and the frequencies corresponding to the first n grids and the grids with frequencies greater than 0.9 are selected. The remaining grid data are removed; the four-stage filtered dataset is obtained, and the abnormal data of the wind turbine is identified.

[0113] This embodiment also discloses a wind turbine abnormal data identification system 200 under complex operating conditions, such as... Figure 2 As shown, it includes:

[0114] Acquisition unit 201 is configured as follows:

[0115] Used to obtain wind turbine ledger data and wind turbine operation data;

[0116] The primary filter unit 202 is configured as follows:

[0117] This is used to filter the wind turbine ledger data and wind turbine operation data based on the wind turbine operating principle, remove shutdown status data, and obtain a first-filtered dataset.

[0118] The secondary filter unit 203 is configured as follows:

[0119] A method for identifying constant value anomaly data based on continuous difference calculation processes the primary filtered data to remove constant value anomaly data and obtain a secondary filtered dataset.

[0120] The third-stage filtration unit 204 is configured as follows:

[0121] The method for identifying power-limited anomaly data processes the secondary filtered data to remove the power-limited anomaly data, resulting in a tertiary filtered dataset.

[0122] The fourth-stage filtering and output unit 205 is configured as follows:

[0123] The method for identifying abnormal data based on grid distribution processes the three-stage filtered dataset to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data.

[0124] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of the wind turbine abnormal data identification method under complex operating conditions.

[0125] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the wind turbine abnormal data identification method under complex operating conditions described in the above embodiments.

[0126] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0127] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0128] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0129] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0130] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0131] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for identifying abnormal data of a wind turbine under complex operating conditions, characterized in that, Includes the following steps: Obtain wind turbine ledger data and wind turbine operation data; Based on the wind turbine operating principle, the wind turbine ledger data and wind turbine operating data are filtered to remove shutdown status data, resulting in a first-filtered dataset. The constant value anomaly data identification method based on continuous difference calculation processes the first filtered data to remove constant value anomalies and obtain a second filtered dataset. Based on the identification method for power limit anomaly data, the secondary filtered data is processed to remove the power limit anomaly data, resulting in a tertiary filtered dataset. The grid-based abnormal data identification method processes the three-stage filtered dataset to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data. The wind turbine ledger data includes the wind turbine cut-in wind speed. Cut-off wind speed Rated wind speed and rated power ; The anomaly data identification method based on grid distribution processes the three-stage filtered dataset to remove scattered anomaly data from the surrounding areas, resulting in a four-stage filtered dataset as follows: Set the average wind speed as the x-axis and the average generator active power as the y-axis, and divide the range of average wind speed by 0.25 intervals. Divide the generator into equal intervals; and use 100 units as the interval for the average range of generator active power. The intervals are divided equally to form a wind speed-generator active power grid; Each data point in the three filtered datasets is mapped to the wind speed-generator active power grid, and the frequency of data contained in each grid is calculated respectively; Filter out all data in the grid whose frequency is less than or equal to 5; Select the average active power range of each generator in sequence, and in each range, obtain the corresponding grid frequency, and calculate the frequency of the generator in the active power range; Sort the frequencies in descending order and select the top ones. The data for each grid cell with a frequency greater than 0.9 are discarded. The average wind speed range of each generator is selected in sequence, and the grid frequency corresponding to each range is obtained. The frequency of the generator in the average wind speed range is calculated. Sort the frequencies in descending order and select the top ones. The data for each grid cell with a frequency greater than 0.9 are discarded. The four filtered datasets were obtained to complete the identification of abnormal wind turbine data.

2. The method for identifying abnormal fan data under complex operating conditions according to claim 1, characterized in that, The wind turbine operating data consists of aggregated average data of the wind turbine at 10-minute intervals over three months, including: time, wind speed, generator active power, pitch angle #1, pitch angle #2, and pitch angle #3.

3. The method for identifying abnormal fan data under complex operating conditions according to claim 1, characterized in that, The process of filtering the wind turbine ledger data and wind turbine operation data based on the wind turbine operating principle to remove shutdown status data and obtain a filtered dataset is as follows: Set the shutdown status data filtering conditions, which include: Wind speed value at cut-in wind speed Cut-off air velocity Outside the range; The average active power of the generator is between 0 and 1.1 times its rated power. Outside the range; The average value of pitch angle 1, pitch angle 2 and pitch angle 3 is greater than 25.

4. The method for identifying abnormal fan data under complex operating conditions according to claim 1, characterized in that, The constant value anomaly data identification method based on continuous difference calculation processes the primary filtered data to remove constant value anomalies and obtain the secondary filtered dataset as follows: Centralized time of a single filtered dataset The dataset is sorted in chronological order. average wind speed Solve the 2nd, 3rd and 4th order backward difference values ​​in sequence to obtain the results of the wind speed difference values ​​for each order. For generator active power The second, third, and fourth order backward difference values ​​are solved sequentially to obtain the results of the generator active power difference values ​​for each order. If there exists a data point where the 2nd, 3rd, and 4th order backward difference values ​​of wind speed or generator active power are all zero, then the current data is identified as a constant value outlier, and the corresponding current data identified as a constant value outlier is deleted, resulting in a secondary filtered dataset.

5. The method for identifying abnormal fan data under complex operating conditions according to claim 1, characterized in that, The method for identifying power-limited anomaly data involves processing the secondary filtered data to remove power-limited anomaly data, resulting in a tertiary filtered dataset. In the secondary filtering dataset, the wind speed is selected based on the cut-in wind speed. and cut-in wind speed Data sets of intervals; Obtain the pitch angles 1#, 2#, and 3# from the filtered dataset and form a pitch angle set. Filter the pitch angle set again to select data in the interval between the 20th and 80th percentiles, and calculate the mean and standard deviation of the pitch angles in this interval. If the active power of a generator in a certain data point is less than 0.9 times the rated power, and the average value of any pitch angle 1, pitch angle 2, and pitch angle 3 is greater than the sum of the average pitch angle and the standard deviation, then the current data is identified as power-limited abnormal data, and the current data corresponding to the data identified as power-limited abnormal data is deleted, resulting in a secondary filtered dataset.

6. The method for identifying abnormal fan data under complex operating conditions according to claim 5, characterized in that, The formula for the method of identifying power-limited abnormal data is expressed as follows: in, For a dataset with three filters, This is a dataset that has undergone secondary filtering. Rated power, This is the average pitch angle. Standard deviation In this context, n represents the nth propeller of the motor, and i represents the time.

7. A system for identifying abnormal data of a wind turbine under complex operating conditions, characterized in that, The method for identifying abnormal wind turbine data under complex operating conditions according to any one of claims 1-6 includes: The acquisition unit is configured as follows: Used to obtain wind turbine ledger data and wind turbine operation data; The primary filter unit is configured as follows: This is used to filter the wind turbine ledger data and wind turbine operation data based on the wind turbine operating principle, remove shutdown status data, and obtain a first-filtered dataset. The secondary filtration unit is configured as follows: A method for identifying constant value anomaly data based on continuous difference calculation processes the primary filtered data to remove constant value anomaly data and obtain a secondary filtered dataset. The three-stage filtering unit is configured as follows: The method for identifying power-limited anomaly data processes the secondary filtered data to remove the power-limited anomaly data, resulting in a tertiary filtered dataset. The four-stage filtering and output unit is configured as follows: The method for identifying abnormal data based on grid distribution processes the three-stage filtered dataset to remove scattered abnormal data from the surrounding area, resulting in a four-stage filtered dataset, thus completing the identification of abnormal wind turbine data.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the wind turbine abnormal data identification method under complex operating conditions as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the wind turbine abnormal data identification method under complex operating conditions as described in any one of claims 1-6.

Images