Wind turbine generator system fault judgment method based on Relief algorithm and related device

Abstract

The invention belongs to the technical field of fault judgment, and discloses a wind turbine generator system fault judgment method based on a Relief algorithm and a related device. The wind turbine generator system fault judgment method based on the Relief algorithm comprises the steps of selecting characteristic quantities and preprocessing the characteristic quantities, iteratively calculating the weight of the preprocessed characteristic quantities according to the Relief algorithm to obtain a characteristic variable set, establishing a regression model between the average wind speed and the characteristic variable set and calculating the regression model, obtaining a normal reference characteristic quantity state of the wind turbine generator system under the real-time state data, and judging whether the wind turbine generator system has a fault or not through a state judgment algorithm according to the real-time state data and the normal reference characteristic quantity state of the wind turbine generator system under the real-time state data; the abnormal condition of the wind turbine generator system can be found in time, an early warning mechanism is triggered, the early warning accuracy is remarkably improved, and occurrence or expansion of faults is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a wind turbine system fault diagnosis method based on a Relief algorithm and related devices. Background Art

[0002] As a vital component of renewable energy, wind turbine systems have been widely adopted worldwide in recent years. With continuous technological advancements and gradual cost reductions, wind turbines have become a key means for many countries to achieve energy transition and reduce carbon emissions. These systems, primarily composed of blades, generators, gearboxes, towers, and other key components, capture wind energy and convert it into electricity.

[0003] Wind turbines are usually installed in remote areas with harsh operating environments, such as high temperature, low temperature, high humidity, strong wind, dust, etc. These environmental factors will affect the normal operation of the unit and increase the difficulty of fault detection. In addition, the structure of wind turbines is complex and consists of multiple components. Each component may fail, and the types of failures are diverse, ranging from electrical failures to mechanical failures, from minor wear and tear to serious damage.

[0004] However, today's fault detection still has the problem of inaccuracy, which may cause more serious equipment damage, lead to safety accidents, and threaten the safety of personnel and equipment. Summary of the Invention

[0005] The purpose of the present invention is to provide a wind turbine system fault diagnosis method and related devices based on the Relief algorithm to overcome the problems existing in the prior art. The present invention can promptly detect abnormal conditions in the wind turbine system and trigger an early warning mechanism, significantly improving the accuracy of the early warning and avoiding the occurrence or expansion of faults.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a wind turbine system fault diagnosis method based on a Relief algorithm, comprising the following steps: Select characteristic quantities based on the monitoring data set of the historical system SCADA and pre-process the characteristic quantities; Iteratively calculate the weights of the pre-processed feature quantities according to the Relief algorithm to obtain a set of feature variables; Establish a regression model between average wind speed and a set of characteristic variables; Obtain real-time status data based on the wind turbine SCADA monitoring data set, calculate the average value of the real-time status data, and input the average value of the real-time status data into the regression model; The regression model is calculated after the average value of the real-time state data is input to obtain the normal reference characteristic quantity state of the wind turbine system under the real-time state data; Based on the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data, a state discrimination algorithm is used to determine whether the wind turbine system has a fault; Furthermore, the pre-processing of the feature quantity specifically includes: cleaning the feature quantity and performing data standardization processing; Furthermore, the iterative calculation of the weight of the pre-processed feature quantity according to the Relief algorithm specifically includes: Take the feature quantity as several training samples, assign weighted initial values ​​to the initial parameters of each dimension of several training samples, and select any one training sample from the several training samples. , respectively find an arbitrarily selected training sample The nearest samples with the same running status and the nearest samples with different operating states ,judge and Are they normal and fault operation state samples respectively? and Calculate the weight value and get the weight of the feature quantity j ; Furthermore, the judgment and Whether it is a normal operating status sample, including: If any training sample is selected If the unit is in normal operating condition, and They are normal and fault operation state samples respectively; If any training sample is selected It is not the normal operating state of the unit. This is not a normal operating status sample. It is not a sample of faulty operating state; Furthermore, the calculation formula of the Relief algorithm is specifically as follows: ; Where, Indicates the i At +1 iteration, the weight j The current value of Indicates the i At the iteration, the weight j The current value of Indicates the target x ′, an arbitrarily selected training sample and The difference between Indicates the target An arbitrarily selected training sample and The difference between n Indicates the number of training samples; Furthermore, the average value of the real-time status data specifically includes: wind speed, wind direction, temperature, power generation, main shaft speed, generator speed; Furthermore, the determining whether there is a fault in the wind turbine system specifically includes: If the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data exceeds a preset threshold interval, then there is a fault in the wind turbine system; If the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data does not exceed the preset threshold range, then there is no fault in the wind turbine system; In a second aspect, the present invention provides a wind turbine system fault diagnosis system based on a Relief algorithm, comprising: The feature quantity selection module is used to select feature quantities based on the monitoring data set of the historical system SCADA and pre-process the feature quantities; The feature variable set acquisition module is used to iteratively calculate the weight of the pre-processed feature quantity according to the Relief algorithm to obtain the feature variable set; A regression model building module is used to build a regression model between the average wind speed and the characteristic variable set; A real-time status data acquisition module is used to acquire real-time status data based on the wind turbine SCADA monitoring data set, calculate the average value of the real-time status data, and input the average value of the real-time status data into the regression model; A regression model calculation module is used to calculate the regression model after inputting the average value of the real-time status data to obtain the normal reference characteristic quantity state of the wind turbine system under the real-time status data; The fault state judgment module is used to judge whether there is a fault in the wind turbine system through a state discrimination algorithm based on the real-time state data and the normal reference characteristic state of the wind turbine system under the real-time state data.

[0007] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0008] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0009] The above technical solution has the following advantages or beneficial effects: In a first aspect, the present invention provides a wind turbine system fault diagnosis method based on the Relief algorithm. By selecting feature quantities from the monitoring data set of the historical system SCADA and performing preprocessing, redundant information can be removed and the features most relevant to the wind turbine system fault can be retained. The weights of the feature quantities are iteratively calculated using the Relief algorithm to obtain a set of feature variables reflecting the importance of the features, which helps to pay more attention to those features that have a significant impact on system faults in subsequent analysis. By establishing a regression model between the average wind speed and the feature variable set, the normal state of the wind turbine at a specific wind speed can be more accurately predicted, thereby providing a more reliable reference for fault diagnosis. Real-time status data can be obtained based on the monitoring data set of the wind turbine SCADA, and the average value of the real-time status data can be calculated. After the average value of the real-time status data is input into the regression model, a normal reference feature quantity state of the wind turbine system under the real-time state can be obtained. By comparing the real-time status data with the reference feature quantity state, abnormal conditions of the wind turbine system can be discovered in a timely manner and an early warning mechanism can be triggered, significantly improving the accuracy of the early warning and avoiding the occurrence or expansion of faults.

[0010] Furthermore, feature cleaning can remove outliers, missing values, or duplicate values ​​in the data to ensure the accuracy of the data for subsequent analysis. Data standardization can convert feature quantities of different dimensions and value ranges to the same scale, making them comparable in the model, thereby improving the stability and accuracy of the model.

[0011] Furthermore, through iterative calculations, the Relief algorithm can accurately evaluate the importance of each feature quantity in distinguishing between normal and faulty operating states of the wind turbine system; the high or low weight value directly reflects the contribution of the feature quantity in fault judgment, providing an important basis for the subsequent establishment of regression models and fault discrimination; by considering the nearest samples of different operating states in the training samples, the Relief algorithm can simulate various operating conditions and fault modes in actual operation to a certain extent.

[0012] Furthermore, by directly The operating status is determined (nearest samples with the same operating status) and The running status of the nearest sample with different running status ensures and The selection of meets the requirements of the Relief algorithm and provides an accurate basis for subsequent weight calculation and fault judgment.

[0013] Furthermore, the average value of real-time status data covers meteorological conditions such as wind speed and wind direction, as well as unit operating parameters such as power generation power, main shaft speed, and generator speed. It can comprehensively reflect the operating status of the wind turbine at a certain moment, help capture subtle changes in unit operation, and provide a rich information basis for fault diagnosis.

[0014] Furthermore, when the difference between the real-time status data and the normal reference characteristic quantity status exceeds the threshold range, the system can issue an alarm in time to prompt the staff to conduct fault troubleshooting and processing, thereby improving the accuracy and reliability of fault judgment; it not only improves the judgment efficiency, but also reduces the possibility of human intervention and misjudgment, providing strong support for the intelligent operation and maintenance of wind turbine systems.

[0015] In the second aspect, the present invention provides a wind turbine system fault judgment system based on the Relief algorithm. The system realizes the systematization and automation of wind turbine system fault judgment by integrating multiple modules such as feature quantity selection, preprocessing, feature variable set acquisition, regression model establishment, real-time status data acquisition, regression model calculation and fault status judgment; the feature quantity selection module can scientifically and rationally select feature quantities that have an important impact on fault judgment based on the monitoring data set of the historical system SCADA, and perform preprocessing; the feature variable set acquisition module uses the Relief algorithm to iteratively calculate the weight of the preprocessed feature quantity to ensure that the selected feature variable set can accurately reflect the operating status of the wind turbine system; the regression model establishment module By establishing a regression model between the average wind speed and the characteristic variable set, the normal operating status of the wind turbine system can be accurately predicted; the real-time status data acquisition module can obtain the operating status data of the wind turbine in real time based on the monitoring data set of the wind turbine SCADA and calculate its average value; the regression model calculation module inputs these real-time status data into the regression model to obtain the normal reference characteristic quantity state of the wind turbine system; the fault status judgment module compares the real-time status data with the reference characteristic quantity state and uses the status discrimination algorithm to quickly judge whether there is a fault in the wind turbine system; the present invention helps to reduce the impact of faults on the operation of the wind turbine and improve the reliability and stability of the system.

[0016] In a third aspect, the present invention provides a computer device that can efficiently implement the steps of the method of the present invention by executing a specific computer program through a processor. When performing data processing tasks, the computer device can accurately perform numerical calculations and logical judgments, avoiding errors caused by human factors; at the same time, since the computer program has a high degree of stability and reliability, the accuracy and consistency of the data processing results can be ensured.

[0017] In a fourth aspect, the present invention provides a computer-readable storage medium. By programming the steps of the method of the present invention into a computer program and storing it on a computer-readable storage medium, users can easily load these programs onto any compatible computer device and execute them without rewriting or converting the code, thereby greatly improving the convenience and flexibility of program execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of a flow chart of a wind turbine system fault diagnosis method based on the Relief algorithm of the present invention; Figure 2 Schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further described in detail below with reference to specific embodiments, which are intended to explain the present invention rather than to limit it. In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention. It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0020] Example: See also Figure 1 The present invention provides a wind turbine system fault judgment method based on the Relief algorithm, comprising the following steps: Step 1: Select feature quantities based on the historical system SCADA monitoring data set, clean the feature quantities, and perform data standardization. Feature cleaning can remove outliers, missing values, or duplicate values ​​in the data to ensure the accuracy of the data for subsequent analysis. Data standardization can convert feature quantities of different dimensions and value ranges to the same scale, making them comparable in the model, thereby improving the stability and accuracy of the model. Step 2: The Relief algorithm iteratively calculates the weights of the preprocessed features to obtain a set of feature variables. Through iterative calculations, the Relief algorithm can accurately assess the importance of each feature in distinguishing between normal and faulty operating states of the wind turbine system. The weight value directly reflects the contribution of the feature to fault diagnosis, providing an important basis for subsequent regression model establishment and fault identification. By considering the nearest neighbor samples of different operating states in the training samples, the Relief algorithm can simulate various operating conditions and fault modes in actual operation to a certain extent. Specifically, the weight of the feature quantity after preprocessing is calculated iteratively according to the Relief algorithm, which includes: taking the feature quantity as a number of training samples, assigning initial weight values ​​to the initial parameters of each dimension of the training samples, and randomly selecting a training sample from the training samples. , respectively find an arbitrarily selected training sample The nearest samples with the same running status and the nearest samples with different operating states , if any training sample is selected If the unit is in normal operating condition, and They are normal and fault operation state samples respectively; if any training sample is selected It is not the normal operating state of the unit. This is not a normal operating status sample. It is not a faulty operating state sample. According to the initial selection parameters of each dimension and the normal operating state sample, and Calculate the weight value and get the weight of the feature quantity j ; Specifically, the calculation formula of the Relief algorithm is: ; Where, Indicates the i At +1 iteration, the weight j The current value of Indicates the i At the iteration, the weight j The current value of Indicates the target x′, an arbitrarily selected training sample and The difference between Indicates the target An arbitrarily selected training sample and The difference between n Indicates the number of training samples; Step 3: Establish a regression model between the average wind speed and the characteristic variable set; Step 4: obtaining real-time status data based on the wind turbine SCADA monitoring data set, calculating the average value of the real-time status data, and inputting the average value of the real-time status data into the regression model; Preferably, the average value of the real-time status data specifically includes: wind speed, wind direction, temperature, power generation, main shaft speed, and generator speed. The average value of the real-time status data covers meteorological conditions such as wind speed and wind direction, as well as unit operating parameters such as power generation, main shaft speed, and generator speed. It can fully reflect the operating status of the wind turbine at a certain moment, help capture subtle changes in unit operation, and provide a rich information basis for fault diagnosis. Step 5: Calculate the average value of the real-time state data input into the regression model to obtain a normal reference characteristic state of the wind turbine system under the real-time state data; Step six, based on the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data, determine whether the wind turbine system has a fault through a status discrimination algorithm. If the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data exceeds a preset threshold interval, then the wind turbine system has a fault; if the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data does not exceed the preset threshold interval, then the wind turbine system does not have a fault.

[0021] In one embodiment of the present invention, a wind turbine system fault diagnosis system based on a Relief algorithm is provided, comprising: The feature quantity selection module is used to select feature quantities based on the monitoring data set of the historical system SCADA and pre-process the feature quantities; The feature variable set acquisition module is used to iteratively calculate the weight of the pre-processed feature quantity according to the Relief algorithm to obtain the feature variable set; A regression model building module is used to build a regression model between the average wind speed and the characteristic variable set; A real-time status data acquisition module is used to acquire real-time status data based on the wind turbine SCADA monitoring data set, calculate the average value of the real-time status data, and input the average value of the real-time status data into the regression model; A regression model calculation module is used to calculate the regression model after inputting the average value of the real-time status data to obtain the normal reference characteristic quantity state of the wind turbine system under the real-time status data; The fault state judgment module is used to judge whether there is a fault in the wind turbine system through a state discrimination algorithm based on the real-time state data and the normal reference characteristic state of the wind turbine system under the real-time state data.

[0022] See also Figure 2 In one embodiment of the present invention, a computer device is provided, comprising a processor and a memory, wherein the memory is configured to store a computer program, the computer program including program instructions, and the processor is configured to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor is the computing core and control core of a terminal and is suitable for implementing one or more instructions, specifically, loading and executing one or more instructions in a computer storage medium to implement a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used to operate a wind turbine system fault diagnosis method based on a relief algorithm.

[0023] In one embodiment of the present invention, a computer-readable storage medium is provided, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device, used to store programs and data. It is understood that the computer-readable storage medium herein may include both built-in storage media in the computer device and, of course, 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 being loaded and executed by a processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer-readable storage medium herein may be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor may load and execute the one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the wind turbine system fault diagnosis method based on the Relief algorithm described in the embodiment.

[0024] Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0025] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate the instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0026] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0027] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A wind turbine system fault diagnosis method based on the Relief algorithm, characterized in that: The following steps are involved: Select characteristic quantities based on the monitoring data set of the historical system SCADA and pre-process the characteristic quantities; Iteratively calculate the weights of the pre-processed feature quantities according to the Relief algorithm to obtain a set of feature variables; Establish a regression model between average wind speed and a set of characteristic variables; Obtain real-time status data based on the wind turbine SCADA monitoring data set, calculate the average value of the real-time status data, and input the average value of the real-time status data into the regression model; The regression model is calculated after the average value of the real-time state data is input to obtain the normal reference characteristic quantity state of the wind turbine system under the real-time state data; According to the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data, a state discrimination algorithm is used to determine whether the wind turbine system has a fault.

2. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 1, characterized in that: The preprocessing of the feature quantity specifically includes: cleaning the feature quantity and data standardization.

3. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 1, characterized in that: The iterative calculation of the pre-processed feature weights according to the Relief algorithm specifically includes: Take the feature quantity as several training samples, assign weighted initial values ​​to the initial parameters of each dimension of several training samples, and select any one training sample from the several training samples. , respectively find an arbitrarily selected training sample The nearest samples with the same running status and the nearest samples with different operating states ,judge and Are they normal and fault operation state samples respectively? and Calculate the weight value and get the weight of the feature quantity j .

4. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 3, characterized in that: The judgment and Whether it is a normal operating status sample, including: If any training sample is selected If the unit is in normal operating condition, and They are normal and fault operation state samples respectively; If any training sample is selected It is not the normal operating state of the unit. This is not a normal operating status sample. Not a faulty operating state sample.

5. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 3, characterized in that: The calculation formula of the Relief algorithm is specifically: ； Where, Indicates the i At +1 iteration, the weight j The current value of Indicates the i At the iteration, the weight j The current value of Indicates the target x ′, an arbitrarily selected training sample and The difference between Indicates the target An arbitrarily selected training sample and The difference between n Indicates the number of training samples.

6. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 1, characterized in that: The average value of the real-time status data specifically includes: wind speed, wind direction, temperature, power generation power, main shaft speed, and generator speed.

7. The wind turbine system fault diagnosis method based on the Relief algorithm according to claim 1, characterized in that: The determining whether the wind turbine system has a fault specifically includes: If the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data exceeds a preset threshold interval, then there is a fault in the wind turbine system; If the difference between the real-time status data and the normal reference characteristic quantity state of the wind turbine system under the real-time status data does not exceed the preset threshold range, there is no fault in the wind turbine system.

8. A wind turbine system fault diagnosis system based on the Relief algorithm, characterized in that: include: The feature quantity selection module is used to select feature quantities based on the monitoring data set of the historical system SCADA and pre-process the feature quantities; The feature variable set acquisition module is used to iteratively calculate the weight of the pre-processed feature quantity according to the Relief algorithm to obtain the feature variable set; A regression model building module is used to build a regression model between the average wind speed and the characteristic variable set; A real-time status data acquisition module is used to acquire real-time status data based on the wind turbine SCADA monitoring data set, calculate the average value of the real-time status data, and input the average value of the real-time status data into the regression model; A regression model calculation module is used to calculate the regression model after inputting the average value of the real-time status data to obtain the normal reference characteristic quantity state of the wind turbine system under the real-time status data; The fault state judgment module is used to judge whether there is a fault in the wind turbine system through a state discrimination algorithm based on the real-time state data and the normal reference characteristic state of the wind turbine system under the real-time state data.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images