Wind turbine generator health state assessment method, system and device and storage medium

Abstract

The invention discloses a wind turbine generator health state assessment method, system and device, and a storage medium. The method comprises the following steps: 1, obtaining historical operation data of a wind turbine generator; step 2, carrying out data screening and reconstruction on the obtained historical operation data; 3, performing power short-term prediction on the wind turbine generator based on an ISSA-RBF neural network and an ARIMA short-term combination prediction model; 4, predicting the health state of the wind turbine generator based on the RHI wind turbine generator health state prediction framework; and 5, displaying a prediction result through a display. The wind turbine generator is predicted from the three aspects of wind power historical data screening and reconstruction, wind power short-term prediction and wind turbine generator health state prediction, the decline trend and the health state of the wind turbine generator are accurately judged in advance according to the prediction result, overhaul and maintenance of equipment can be reasonably arranged, the normal operation quality of the wind turbine generator is improved, and the working efficiency is improved. And the post-maintenance cost of the fault is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid user load regulation, and in particular to a wind turbine health state evaluation method, system, device and storage medium. BACKGROUND

[0002] In recent years, wind power has developed rapidly, but the grid connection of wind power is unstable, resulting in low utilization rate of wind power. Wind power plants are located in remote areas, and it is difficult to operate and maintain. At present, wind power plants still adopt the method of post-maintenance and periodic maintenance, and the maintenance cost is high during operation.

[0003] Due to the instability and high randomness of wind energy itself, the control of wind power generation is more difficult. The uncontrollability of wind power integration has a great impact on the traditional power grid, and seriously affects the ability of the power grid to absorb wind power. At present, the power grid in China cannot timely absorb a large amount of wind energy. In order to ensure the reliable and stable operation of wind power integrated into the power grid, the power grid dispatching department must take a series of measures, for example, reduce the installed capacity of wind power generation. Limiting the output of the wind turbine will also lead to large-scale wind power curtailment in the wind farm, causing unnecessary resource waste and increasing maintenance costs. It will also lead to large-scale wind power curtailment in the wind farm. The maintenance cost of the wind turbine accounts for an increasing proportion of the total cycle cost, and the economically affordable price has become the main concern of companies in different countries and regions, and power companies are no exception.

[0004] If the degradation trend and health status of the wind turbine can be accurately predicted in advance through monitoring data, the maintenance and repair of the equipment can be reasonably arranged, the quality of the normal operation of the unit can be improved, and the post-maintenance cost of the fault can be reduced, which has certain application and practical significance. SUMMARY

[0005] The technical problem solved by the present application is to provide a wind turbine health state evaluation method, system, device and storage medium, which can accurately predict the degradation trend and health status of the wind turbine in advance through monitoring the operation data of the wind turbine, reasonably arrange the maintenance and repair of the equipment, improve the quality of the normal operation of the unit, and reduce the post-maintenance cost of the fault.

[0006] To solve the above technical problems, one technical solution adopted by the present application is to provide a wind turbine health state evaluation method, characterized by comprising the following steps:

[0007] Step 1, obtaining historical operation data of the wind turbine;

[0008] Step 2, data screening and reconstruction of the obtained historical operation data;

[0009] Step three, the short-term power prediction of the wind turbine is performed based on the short-term combination prediction model of the ISSA-RBF neural network and the ARIMA;

[0010] Step four, the health state of the wind turbine is predicted based on the wind turbine health state prediction framework of the RHI;

[0011] Step five, the prediction result is displayed through a display.

[0012] Further, in the step one, the operation data of the wind turbine includes environmental data and working data;

[0013] The environmental data is the temperature, humidity, wind force and wind direction of the location of the wind turbine;

[0014] The working data is the construction time, working time, maintenance record, vibration record, voltage record, current record, temperature record and generator speed record.

[0015] Further, in the step one, the process of obtaining the historical operation data of the wind turbine is that the data acquisition port is connected with the control system of the wind turbine, the operation data of the wind turbine is obtained in real time by using the sensor of the wind turbine, and the operation data is classified and stored according to the date and type.

[0016] Further, in the step two, the process of data screening and reconstruction is that 1) the historical data is classified according to the reasons for data generation;

[0017] 2) for the wind curtailment and wind limit data, the power curve method is used for reconstruction;

[0018] 3) for the outlier data, the quartile method is used for identification, and the interpolation method is used for reconstruction;

[0019] 4) the accuracy of the reconstructed data is quantitatively analyzed by taking the average relative error and the accuracy as the basis for judgment.

[0020] Further, in the step three, the process of the short-term power prediction of the wind turbine is that

[0021] 1) the data is decomposed into low-frequency and high-frequency components by using the variational mode decomposition algorithm;

[0022] 2) the ISSA-RBF model is used to predict the low-frequency component;

[0023] 3) the ARIMA model which is suitable for non-stationary signals is used to process the high-frequency component;

[0024] 4) the prediction results of each component are reconstructed to obtain the final prediction result.

[0025] Further, the step four is to predict the health state of the wind turbine:

[0026] 1) using a short-term combined prediction model to obtain a standard residual set of wind turbine parameters and a real-time residual set under actual operation state;

[0027] 2) using Mahalanobis distance to describe the similarity of the two data sets, and obtaining the RHI value of the wind turbine after normalization processing;

[0028] 3) obtaining the health state change trend of the wind turbine according to the obtained RHI value.

[0029] To solve the above technical problems, another technical solution adopted by the present application is to provide a wind turbine health state evaluation system, characterized by comprising a data acquisition module, a data processing module, a state evaluation module and a display module, wherein:

[0030] The data acquisition module is connected with the control system of the wind turbine through a data acquisition port, and real-time operation data of the wind turbine is obtained by using the sensors of the wind turbine, and the historical operation data of the wind turbine is classified and stored according to date and type;

[0031] The data processing module classifies the historical data collected by the data acquisition module according to the reasons for data generation; for wind curtailment and wind limit data, the data is reconstructed according to the power curve method; for outlier data, the data is identified by using the quartile method and reconstructed according to the interpolation method; the accuracy of the reconstructed data is quantitatively analyzed by taking the average relative error and accuracy as the basis; and the historical operation data obtained is screened and reconstructed;

[0032] The state evaluation module retrieves the data processed by the data processing module, performs short-term power prediction on the wind turbine based on the short-term combined prediction model of the ISSA-RBF neural network and ARIMA, predicts the health state of the wind turbine based on the RHI wind turbine health state prediction framework, and sends the prediction information to the display module;

[0033] The display module obtains and displays the prediction results sent by the state evaluation module.

[0034] Further, the state evaluation module comprises a short-term power prediction module and a health state prediction module.

[0035] Short-term power prediction module: Using variational mode decomposition algorithm, the data is decomposed into low-frequency and high-frequency components; for low-frequency components, the ISSA-RBF model is used for prediction; for high-frequency components, the ARI MA model, which is more applicable to non-stationary signals, is used; the prediction results of each component are reconstructed to obtain the final prediction result.

[0036] Health status prediction module: Using a short-term combined prediction model, the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions are obtained; the similarity between the two datasets is described by Mahalanobis distance, and after normalization, the RHI value of the wind turbine is obtained; the trend of the wind turbine's health status change is obtained based on the obtained RHI value.

[0037] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is: to provide a wind turbine health status assessment device, characterized in that it includes:

[0038] Memory, used to store computer programs;

[0039] A processor is configured to read and execute the computer program stored in the memory, wherein when the computer program is executed, the processor executes a wind turbine health status assessment method as described above.

[0040] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is to provide a computer-readable storage medium, characterized in that: the computer-readable storage medium stores instructions, which, when the computer instructions are executed on a computer, cause the computer to execute the wind turbine health status assessment method described in any of the above-mentioned claims.

[0041] The beneficial effects of this invention are:

[0042] 1. This invention enables the prediction of wind turbines from three aspects: historical wind power data screening and reconstruction, short-term wind power prediction, and wind turbine health status prediction. Based on the prediction results, the decline trend and health status of wind turbines can be accurately judged in advance, so as to reasonably arrange equipment inspection and maintenance, improve the quality of normal operation of the unit, and reduce the cost of post-failure maintenance.

[0043] 2. This invention provides more accurate prediction of wind turbine power, which can reduce the impact of wind power grid connection on the stability of the main power grid, enable more scientific grid dispatching, and thus effectively improve wind power utilization. It also provides health prediction of turbine operating status, which can more accurately grasp the information on the operating status of wind turbines. This is of great significance for improving the intelligence of wind farm dispatching and control, reducing operation and maintenance costs, and increasing the economic income of wind farms.

[0044] 3. In order to improve the accuracy of prediction, this invention employs a method for filtering and reconstructing historical data, removing erroneous data from historical wind power data, thus solving the problem that directly using historical data for prediction can significantly impact prediction accuracy.

[0045] 4. This invention utilizes a variational mode decomposition algorithm to decompose data into low-frequency and high-frequency components, and employs different models for prediction of different components. Finally, the prediction results of each component are reconstructed to obtain the final prediction result, further improving the prediction accuracy of the model and solving the problem of low short-term prediction accuracy of wind power.

[0046] 5. This invention obtains the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions through power prediction results, and uses Mahalanobis distance to describe the similarity between the two datasets, thus solving the problem of inaccurate judgment of wind turbine operating conditions.

[0047] To make the above and other objects, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only four of the drawings in this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0049] Figure 1 This is a flowchart of the method described in this application;

[0050] Figure 2 This is a structural block diagram of the system in this application;

[0051] Figure 3 This is a structural block diagram of the device in this application;

[0052] Figure 4 This is a structural block diagram of the storage medium of this application. Detailed Implementation

[0053] To facilitate understanding of the present invention, a more detailed description is provided below with reference to the accompanying drawings and specific embodiments. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0054] It should be noted that, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0055] Please see Figure 1 , Figure 1 This invention discloses a method for assessing the health status of a wind turbine generator, characterized by comprising the following steps:

[0056] Step S1: Obtain historical operating data of the wind turbine;

[0057] Step S2: Filter and reconstruct the acquired historical operational data;

[0058] Step S3: Perform short-term power prediction for wind turbines based on the short-term combined prediction model of ISSA-RBF (Improve Sparrow Search Algorithm-Radical BasisFunction) neural network and ARIMA (Auto Regressive Integrated Moving Average);

[0059] Step S4: The wind turbine health status prediction framework based on RHI is used to predict the health status of the wind turbine.

[0060] Step S5: The prediction results are displayed on the monitor.

[0061] In step S1 above, the operating data of the wind turbine includes environmental data and operational data;

[0062] Environmental data includes the temperature, humidity, wind speed, and wind direction at the location of the wind turbine.

[0063] The work data includes construction time, working hours, maintenance records, vibration records, voltage records, current records, temperature records, and generator speed records.

[0064] The process of acquiring historical operating data of wind turbines is as follows: the data acquisition port is connected to the control system of the wind turbine, and the operating data of the wind turbine is acquired in real time by the wind turbine's own sensors and stored according to date and type.

[0065] In step S2 above, the process of data filtering and reconstruction is as follows: 1) Classify historical data according to the reasons for data generation;

[0066] 2) For wind curtailment and wind restriction data, reconstruct them using the power curve method;

[0067] 3) For outlier data, the quartile method is used to identify them and the data is reconstructed using interpolation.

[0068] 4) The accuracy of the reconstructed data is quantitatively analyzed based on the average relative error and accuracy rate.

[0069] In step S3 above, the process of short-term power prediction for wind turbine units is as follows:

[0070] 1) Using variational mode decomposition algorithm, the data is decomposed into low-frequency and high-frequency components;

[0071] 2) For low-frequency components, the ISSA-RBF model is used for prediction;

[0072] 3) For high-frequency components, the ARIMA model, which is well-suited for non-stationary signals, is used for processing.

[0073] 4) The prediction results of each component were reconstructed to obtain the final prediction results.

[0074] In step S4 above, the process of predicting the health status of the wind turbine is as follows:

[0075] 1) Using a short-term combined prediction model, obtain the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions;

[0076] 2) The similarity between two datasets is described by Mahalanobis distance, and the RHI value of the wind turbine is obtained after normalization.

[0077] 3) The health status trend of the wind turbine unit can be determined based on the obtained RHI value.

[0078] Please see Figure 2 , Figure 2 This application provides a wind turbine health status assessment system 2, characterized in that it includes a data acquisition module 21, a data processing module 22, a status assessment module 23, and a display module 24, wherein:

[0079] Data acquisition module 21: Connects to the control system of the wind turbine through the data acquisition port, uses the wind turbine's own sensors to acquire the wind turbine's operating data in real time, classifies and stores it according to date and type, and acquires the wind turbine's historical operating data;

[0080] Data processing module 22: Classifies the historical data acquired by the data acquisition module according to the reasons for data generation; reconstructs the wind curtailment and wind restriction data according to the power curve method; identifies outlier data using the quartile method and reconstructs it according to the interpolation method; quantitatively analyzes the accuracy of the reconstructed data based on the average relative error and accuracy rate; thereby filtering and reconstructing the acquired historical operation data.

[0081] State assessment module 23: retrieves data processed by the data processing module, performs short-term power prediction of wind turbines based on the short-term combined prediction model of ISSA-RBF (ImproveSparrow Search Algorithm-Radical Basis Function) neural network and ARIMA (AutoRegressive Integrated Moving Average); predicts the health status of wind turbines based on the RHI wind turbine health status prediction framework; and sends the prediction information to the display module.

[0082] Display module 24: Obtains and displays the prediction results sent by the status assessment module.

[0083] Furthermore, the status assessment module 23 includes a short-term power prediction module 231 and a health status prediction module 232;

[0084] Power short-term prediction module 231: Using variational mode decomposition algorithm, the data is decomposed into low-frequency and high-frequency components; for the low-frequency components, the ISSA-RBF model is used for prediction; for the high-frequency components, the ARIMA model, which is more applicable to non-stationary signals, is used; the prediction results of each component are reconstructed to obtain the final prediction result.

[0085] Health status prediction module 232: Using a short-term combined prediction model, the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions are obtained; the similarity between the two datasets is described by Mahalanobis distance, and the RHI value of the wind turbine is obtained after normalization; the trend of the wind turbine's health status change is obtained based on the obtained RHI value.

[0086] Please see Figure 3 , Figure 3 This application provides a wind turbine health status assessment device 3, comprising:

[0087] Memory 31 is used to store computer programs.

[0088] The processor 32 is configured to read and execute the computer program stored in the memory. When the computer program is executed, the processor executes any of the wind turbine health status assessment methods described above.

[0089] The processor 32 is used to execute the program instructions stored in the memory 31 to implement the steps of any of the above embodiments of the wind turbine health status assessment method. In a specific implementation scenario, the wind turbine health status assessment device 3 may include, but is not limited to, a microcomputer or a server. In addition, the wind turbine health status assessment device 3 may also include mobile devices such as laptops and tablets, which are not limited here.

[0090] Specifically, processor 32 controls itself and memory 31 to implement the steps of any of the wind turbine health status assessment method embodiments described above. Processor 32 can also be referred to as a CPU (Central Processing Unit). Processor 32 may be an integrated circuit chip with signal processing capabilities. Processor 32 can also be a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor. Furthermore, processor 32 can be implemented using integrated circuit chips.

[0091] Please see Figure 4 , Figure 4 This is a schematic diagram of a framework of an embodiment of the computer-readable storage medium provided in this application. The computer-readable storage medium 4 stores program instructions 41 that can be executed by a processor. The program instructions 41 are used to implement the steps of any of the above embodiments of the wind turbine health status assessment method.

[0092] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0093] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities can be referred to each other. For the sake of brevity, this application will not repeat them.

[0094] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0095] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] The above are merely embodiments of the present invention and are not intended to limit the scope of the patent of the present invention. Any equivalent structural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for assessing the health status of wind turbine generators, characterized in that: Includes the following steps: Step 1: Obtain historical operating data of the wind turbine; Step 2: Filter and reconstruct the acquired historical operational data; Step 3: Perform short-term power prediction for wind turbines using a short-term combined prediction model based on ISSA-RBF neural network and ARIMA. Step 4: Predict the health status of wind turbines using the RHI-based wind turbine health status prediction framework. Step 5: The prediction results are displayed on the monitor.

2. The method for assessing the health status of a wind turbine generator according to claim 1, characterized in that: In step one, the operating data of the wind turbine includes environmental data and operational data; Environmental data includes the temperature, humidity, wind speed, and wind direction at the location of the wind turbine. The work data includes construction time, working hours, maintenance records, vibration records, voltage records, current records, temperature records, and generator speed records.

3. The method for assessing the health status of a wind turbine generator according to claim 1, characterized in that: In step one, the process of acquiring historical operating data of the wind turbine is as follows: the wind turbine is connected to the control system of the wind turbine through the data acquisition port, and the operating data of the wind turbine is acquired in real time by the wind turbine's own sensors and stored according to date and type.

4. The method for assessing the health status of a wind turbine generator according to claim 1, characterized in that: In step two, the process of data filtering and reconstruction is as follows: 1) Classify historical data according to the reasons for data generation; 2) For wind curtailment and wind restriction data, reconstruct them using the power curve method; 3) For outlier data, the quartile method is used to identify them and the data is reconstructed using interpolation. 4) The accuracy of the reconstructed data is quantitatively analyzed based on the average relative error and accuracy rate.

5. The method for assessing the health status of a wind turbine generator according to claim 1, characterized in that: In step three, the process of short-term power prediction for wind turbine units is as follows: 1) Using variational mode decomposition algorithm, the data is decomposed into low-frequency and high-frequency components; 2) For low-frequency components, the ISSA-RBF model is used for prediction; 3) For high-frequency components, the ARIMA model, which is well-suited for non-stationary signals, is used for processing. 4) The prediction results of each component were reconstructed to obtain the final prediction results.

6. The method for assessing the health status of a wind turbine generator according to claim 5, characterized in that: In step four, the process of predicting the health status of the wind turbine is as follows: 1) Using a short-term combined prediction model, obtain the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions; 2) The similarity between two datasets is described by Mahalanobis distance, and the RHI value of the wind turbine is obtained after normalization. 3) The health status trend of the wind turbine unit can be determined based on the obtained RHI value.

7. A wind turbine health status assessment system, characterized in that: It includes a data acquisition module, a data processing module, a status assessment module, and a display module, among which: Data acquisition module: It connects to the wind turbine control system through the data acquisition port, uses the wind turbine's own sensors to acquire the wind turbine's operating data in real time, classifies and stores it according to date and type, and obtains the wind turbine's historical operating data; Data processing module: Classifies historical data acquired by the data acquisition module according to the reasons for data generation; reconstructs wind curtailment and wind restriction data using the power curve method; identifies outlier data using the quartile method and reconstructs it using interpolation; quantitatively analyzes the accuracy of the reconstructed data based on average relative error and accuracy rate; thereby filtering and reconstructing the acquired historical operating data. The status assessment module retrieves data processed by the data processing module, performs short-term power prediction of the wind turbine based on the short-term combined prediction model of ISSA-RBF neural network and ARIMA, predicts the health status of the wind turbine based on the RHI wind turbine health status prediction framework, and sends the prediction information to the display module. Display module: Obtains and displays the prediction results sent by the status assessment module.

8. The wind turbine health status assessment system according to claim 1, characterized in that: The condition assessment module includes a short-term power prediction module and a health status prediction module; Short-term power prediction module: 1) Using variational mode decomposition algorithm, the data is decomposed into low-frequency and high-frequency components; for low-frequency components, the ISSA-RBF model is used for prediction; for high-frequency components, the ARIMA model, which is more applicable to non-stationary signals, is used; the prediction results of each component are reconstructed to obtain the final prediction result. Health status prediction module: Using a short-term combined prediction model, the standard residual set of wind turbine parameters and the real-time residual set under actual operating conditions are obtained; the similarity between the two datasets is described by Mahalanobis distance, and after normalization, the RHI value of the wind turbine is obtained; the trend of the wind turbine's health status change is obtained based on the obtained RHI value.

9. A health status assessment device for wind turbine generators, characterized in that, include: Memory, used to store computer programs; A processor is configured to read and execute the computer program stored in the memory, wherein when the computer program is executed, the processor performs a wind turbine health status assessment method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform a wind turbine health status assessment method as described in any one of claims 1-6.

Images