Hierarchical fault early warning method for wind turbine generator
By providing a layered fault warning for the wind turbine, first monitoring the entire machine, and then positioning it to the abnormal subsystem, the problem of high resource consumption in the existing technology is solved and more efficient monitoring and maintenance is achieved.
Patent Information
- Application Number
- CN202411962218.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-04-18
AI Technical Summary
In the prior art, each subsystem is directly monitored, resulting in high resource consumption, which is not conducive to improving the operating efficiency of the monitoring system.
A hierarchical fault warning method for wind turbines is proposed. By first monitoring the entire machine, when an abnormality occurs, the subsystem that predicts the problem based on the abnormality, and further monitoring the subsystem to quickly locate the abnormal parts.
Through the hierarchical monitoring method, it is possible to quickly locate the abnormal parts, reduce resource consumption, and improve the operating efficiency of the monitoring system.
Smart Images
Figure CN119848732A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of wind turbine fault warning, and in particular relates to a wind turbine hierarchical fault warning method. Background Art
[0002] Wind turbines are a type of power generation equipment that converts wind energy into electrical energy and are the core of wind power generation. The operating status of wind turbines has a significant impact on their power generation efficiency and stable operation of the equipment. Therefore, it is of great significance to effectively collect and analyze the operating data of wind turbines, conduct predictive maintenance through hierarchical fault warning, and improve the operating efficiency of wind turbines.
[0003] In the prior art, each subsystem is directly monitored. Since there are many subsystems to be monitored, a large amount of resources are consumed, which is not conducive to improving the operating efficiency of the monitoring system. Summary of the invention
[0004] The purpose of the present invention is to provide a hierarchical fault early warning method for wind turbine components, so as to solve the problem in the prior art that directly monitoring each subsystem is not conducive to improving the operating efficiency of the monitoring system.
[0005] To this end, the present invention provides a wind turbine component hierarchical fault early warning method, comprising the following steps:
[0006] Acquire SCADA data, pre-process the SCADA data, and obtain pre-processed data;
[0007] Input the preprocessed data into the early warning model to determine whether there is any whole machine abnormality;
[0008] If there is an abnormality in the entire machine, each subsystem will be tested separately through the early warning model to locate the subsystem with the abnormality.
[0009] Preferably, the SCADA data includes historical data and real-time data, the historical data is used to determine the threshold value of the unit health status indicator, and the real-time data is used to determine the real-time status indicator.
[0010] Preferably, the real-time data is collected periodically to form a chronological data sequence.
[0011] Preferably, the historical data records the fault data that have occurred in the entire machine and each subsystem.
[0012] Preferably, the early warning model includes a whole machine early warning model and a subsystem early warning model established for each subsystem.
[0013] Preferably, the whole machine establishes several different algorithm models according to the fault data in the historical data, uses the fault data to test the algorithm models, obtains the early warning success rate and false alarm rate of each algorithm model, and selects the optimal prediction model as the whole machine early warning model according to the early warning success rate and false alarm rate.
[0014] Preferably, the subsystem includes at least: a wind wheel subsystem, a transmission subsystem, and a generator subsystem.
[0015] Preferably, each subsystem predicts the location of anomalies through a subsystem early warning model.
[0016] Preferably, the subsystem early warning model includes three subsystem prediction models.
[0017] Preferably, the subsystem establishes several different algorithm models based on the fault data in the historical data, uses the fault data to test the algorithm models, obtains the early warning success rate and false alarm rate of each algorithm model, and selects the optimal prediction model according to the early warning success rate and false alarm rate as the subsystem early warning model.
[0018] Beneficial effects:
[0019] 1. The present invention provides a hierarchical fault warning method for wind turbines. Through a hierarchical monitoring method, the whole machine is first monitored. When an abnormality occurs, the subsystem with the problem is predicted based on the abnormality. The subsystem is then monitored and the abnormal point is predicted. The abnormal part can be quickly located to reduce resource consumption and improve the operating efficiency of the monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0021] Figure 1 This is a method step diagram of a wind turbine component hierarchical fault early warning method in the present invention.
[0022] Figure 2 The present invention is a method flow chart of a wind turbine component hierarchical fault early warning method.
[0023] Figure 3 The present invention is a flowchart of the early warning steps of a wind turbine component hierarchical fault early warning method. DETAILED DESCRIPTION
[0024] The content of the present invention can be more easily understood by selecting the following detailed description of the preferred implementation method of the present invention and the embodiments included. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those of ordinary skill in the art to which the present invention belongs. When there is a conflict, the definition in this specification shall prevail.
[0025] Embodiment 1:
[0026] like Figure 1-3 As shown, this embodiment provides a wind turbine component hierarchical fault early warning method, including the following steps:
[0027] S1, obtaining SCADA data, preprocessing the SCADA data to obtain preprocessed data;
[0028] SCADA (Supervisory Control and Data Acquisition) is a data acquisition and monitoring control system. SCADA data refers to various data collected through the system.
[0029] Among them, SCADA data includes historical data and real-time data. Historical data is used to determine the threshold of the unit health status indicator, and real-time data is used to determine the real-time status indicator. The historical data records the fault data that has occurred in the whole machine and each subsystem. The real-time data is collected periodically through the data acquisition and monitoring control system to form a chronological data sequence. The historical data and real-time data are pre-processed through data cleaning, noise removal, and standardization.
[0030] The SCADA system can collect and record all-round operation information of wind turbines in real time, including multiple attribute parameters such as ambient wind speed, generator speed, active power, etc. The acquisition system of the unit control center controls the remote sensors to record the unit monitoring data at regular intervals to form a chronological data sequence. Some relevant performance parameters and normal ranges are shown in Table 1.
[0031] Table 1 Some relevant performance parameters of wind turbines and their normal range
[0032]
[0033]
[0034] S2. Input the preprocessed data into the early warning model to determine whether there is an abnormality in the whole machine;
[0035] The early warning model includes the whole machine early warning model and the subsystem early warning model of each subsystem.
[0036] The whole machine establishes several different algorithm models according to the fault data in the historical data, and uses the fault data to test the algorithm models to obtain the early warning success rate and false alarm rate of each algorithm model. The optimal early warning model is selected as the early warning model of the whole machine according to the early warning success rate and false alarm rate.
[0037] The preprocessed data obtained from historical data and real-time data are input into the whole machine prediction model. The historical data obtains the status indicators under the normal state, and the real-time data obtains the status indicators under the monitoring state. The status indicators under the monitoring state are compared with the status indicators under the normal state to determine whether they exceed the threshold. If they do not exceed the threshold, return to S1 to monitor the data again. If they exceed the threshold, perform S3 to detect the subsystem.
[0038] S3. If the whole machine is abnormal, each subsystem is tested separately through the early warning model to locate the subsystem with the abnormality.
[0039] The subsystems include at least: a wind wheel subsystem, a transmission subsystem, and a generator subsystem.
[0040] Each subsystem predicts the location of abnormality through the subsystem early warning model. The subsystem early warning model includes three subsystem prediction models. The subsystem establishes several different algorithm models based on the fault data in the historical data, tests the algorithm model using the fault data, obtains the early warning success rate and false alarm rate of each algorithm model, and selects the optimal prediction model based on the early warning success rate and false alarm rate as the early warning model for the three subsystems.
[0041] The historical data and real-time data are preprocessed to separate the related subsystem preprocessed data, and the subsystem preprocessed data is input into the subsystem prediction model. The historical data obtains the status indicators under the normal state, and the real-time data obtains the status indicators under the monitoring state. The status indicators under the monitoring state are compared with the status indicators under the normal state to determine whether they exceed the threshold. If the threshold is not exceeded, the data of the next subsystem is monitored. If the threshold is exceeded, the abnormal subsystem is located and the cause of the abnormality is predicted.
[0042] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A wind turbine component hierarchical fault early warning method, characterized in that: The steps include: Acquire SCADA data, pre-process the SCADA data, and obtain pre-processed data; Input the preprocessed data into the early warning model to determine whether there is any whole machine abnormality; If there is an abnormality in the entire machine, each subsystem will be tested separately through the early warning model to locate the subsystem with the abnormality.
2. A wind turbine component hierarchical fault early warning method according to claim 1, characterized in that: The SCADA data includes historical data and real-time data, wherein the historical data is used to determine the threshold of the unit health status indicator, and the real-time data is used to determine the real-time status indicator.
3. A wind turbine component hierarchical fault early warning method according to claim 2, characterized in that: The real-time data is collected periodically to form a data sequence in chronological order.
4. A wind turbine component hierarchical fault early warning method according to claim 2, characterized in that: The historical data records the fault data that have occurred in the entire machine and each subsystem.
5. A wind turbine component hierarchical fault early warning method according to claim 4, characterized in that: The early warning model includes a whole machine early warning model and a subsystem early warning model established for each subsystem.
6. A wind turbine component hierarchical fault early warning method according to claim 5, characterized in that: The whole machine establishes several different algorithm models according to the fault data in the historical data, uses the fault data to test the algorithm models, obtains the early warning success rate and false alarm rate of each algorithm model, and selects the optimal early warning model as the whole machine early warning model according to the early warning success rate and false alarm rate.
7. A wind turbine component hierarchical fault early warning method according to claim 1, characterized in that: The subsystems at least include: a wind wheel subsystem, a transmission subsystem, and a generator subsystem.
8. A wind turbine component hierarchical fault early warning method according to claim 7, characterized in that: Each subsystem predicts the location of abnormalities through the subsystem early warning model.
9. A wind turbine component hierarchical fault early warning method according to claim 8, characterized in that: The subsystem early warning model includes early warning models of three subsystems.
10. A wind turbine component hierarchical fault early warning method according to claim 1, characterized in that: The subsystem establishes several different algorithm models based on the fault data in the historical data, uses the fault data to test the algorithm models, obtains the early warning success rate and false alarm rate of each algorithm model, and selects the optimal prediction model based on the early warning success rate and false alarm rate as the subsystem early warning model.