CN-120819475-B - Frequency monitoring-based doubly-fed wind turbine generator fault monitoring method and system

CN120819475BCN 120819475 BCN120819475 BCN 120819475BCN-120819475-B

Abstract

The application relates to the technical field of fault monitoring of wind turbines, in particular to a frequency monitoring-based doubly-fed wind turbine fault monitoring method and a frequency monitoring-based doubly-fed wind turbine fault monitoring system, wherein the method comprises the steps of obtaining operation parameter data of the doubly-fed wind turbine, and generating a similar working condition time period of a gearbox in the doubly-fed wind turbine according to the operation parameter data based on a clustering algorithm; the method comprises the steps of extracting target vibration data, generating modal components and center frequencies corresponding to the modal components according to the target vibration data, generating modal component energy, generating fusion modal energy values according to the modal component energy, and generating a unit fault early warning result. According to the application, only vibration data corresponding to similar working conditions in a time period are extracted, interference of signal difference on vibration characteristics under different working conditions is eliminated, and the fault risk of the gearbox is accurately pre-warned as much as possible and missing report is reduced by carrying out comprehensive criteria of unit faults based on the energy values of fusion modes.

Inventors

MA DEZHI
LI WENYI
KOU ZHIWEI

Assignees

内蒙古工业大学

Dates

Publication Date: 20260508
Application Date: 20250729

Claims (10)

1. The method for monitoring the faults of the doubly-fed wind turbine based on the frequency monitoring is characterized by comprising the following steps of: Acquiring operation parameter data of the doubly-fed wind turbine, and generating a similar working condition time period of a gear box in the doubly-fed wind turbine according to the operation parameter data based on a clustering algorithm; Extracting target vibration data from the vibration monitoring data of the gear box according to the similar working condition time period; Generating a modal component and a center frequency corresponding to each modal component according to the target vibration data, and generating modal component energy according to the modal component and the center frequency corresponding to each modal component; And generating a fusion modal energy value according to the modal component energy, and generating a unit fault early warning result according to the fusion modal energy value.
2. The frequency monitoring-based doubly-fed wind turbine fault monitoring method according to claim 1, wherein obtaining operational parameter data of a doubly-fed wind turbine, generating a similar operating condition time period of a gearbox in the doubly-fed wind turbine according to the operational parameter data based on a clustering algorithm, comprises: Acquiring operation parameter data of the doubly-fed wind turbine generator, and generating a multidimensional working condition feature vector according to the operation parameter data; normalizing and distributing weight to the multidimensional working condition feature vector, and generating a weighted feature vector; segmenting the weighted feature vector according to a preset segmentation interval, and screening out a target working condition segment based on a clustering algorithm and preset screening conditions; And generating a similar working condition time period according to the target working condition time period.
3. The frequency monitoring-based doubly-fed wind turbine generator fault monitoring method according to claim 2, wherein the preset segmented interval comprises a wind speed interval and a pitch angle interval; Segmenting the weighted feature vector according to a preset segmentation interval, and screening out a target working condition segment based on a clustering algorithm and a preset screening condition, wherein the method comprises the following steps: segmenting the weighted feature vector according to the wind speed interval and the pitch angle interval, and respectively clustering the data in each segment by adopting a clustering algorithm to obtain a plurality of working condition clusters; Screening target working condition segments from the working condition clusters according to preset screening conditions; And calculating the center vector of each target working condition segment, and screening the target working condition segments from the target working condition segments according to the center vector.
4. The frequency monitoring-based doubly-fed wind turbine generator fault monitoring method according to claim 1, wherein the vibration monitoring data comprises first vibration data and second vibration data, wherein the first vibration data is acquired based on a first vibration sensor, and the second vibration data is acquired based on a second vibration sensor; extracting target vibration data from the vibration monitoring data of the gearbox according to the similar working condition time period, wherein the extracting target vibration data comprises the following steps: Acquiring first vibration data and second vibration data of the gear box; generating differential vibration data from the first vibration data and the second vibration data; and extracting target vibration data from the differential vibration data according to the similar working condition time period.
5. The frequency monitoring-based doubly-fed wind turbine generator fault monitoring method according to claim 1, wherein generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energy according to the modal components and the center frequencies corresponding to the modal components, comprises: Decomposing the target vibration data based on a variation modal decomposition method, and generating a plurality of modal components and center frequencies corresponding to the modal components; screening out target modal components according to the modal components, the center frequency and the estimated meshing frequency of the main gear; And generating modal component energy according to the target modal component.
6. The frequency monitoring-based doubly-fed wind turbine generator fault monitoring method according to claim 1, wherein generating a fusion modal energy value according to the modal component energy and generating a generator fault early warning result according to the fusion modal energy value comprises: Generating a fusion modal energy value according to the modal component energy; generating an energy growth rate and an energy fluctuation anomaly degree according to the fusion mode energy value corresponding to each target vibration data; And generating a unit fault early warning result according to the energy growth rate and the energy fluctuation abnormality degree.
7. The frequency monitoring-based doubly-fed wind turbine generator fault monitoring method according to claim 6, wherein the generator fault early warning result comprises a generator fault early warning and a generator normal indication; Generating a unit fault early warning result according to the energy growth rate and the energy fluctuation abnormality degree, wherein the unit fault early warning result comprises: judging whether the energy growth rate is greater than the health growth rate or not, and judging that the energy fluctuation anomaly degree is greater than the health anomaly degree; If yes, generating a unit fault early warning; if not, generating a unit normal instruction.
8. A doubly-fed wind turbine generator fault monitoring system based on frequency monitoring, the system comprising: The similar working condition generation module is used for acquiring the operation parameter data of the double-fed wind turbine generator and generating a similar working condition time period of a gear box in the double-fed wind turbine generator based on a clustering algorithm according to the operation parameter data; The vibration data screening module is used for extracting target vibration data from the vibration monitoring data of the gearbox according to the similar working condition time period; The modal energy generation module is used for generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energy according to the modal components and the center frequencies corresponding to the modal components; And the fault early warning generation module is used for generating a fusion modal energy value according to the modal component energy and generating a unit fault early warning result according to the fusion modal energy value.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Description

Frequency monitoring-based doubly-fed wind turbine generator fault monitoring method and system Technical Field The application relates to the technical field of fault monitoring of wind turbines, in particular to a frequency monitoring-based doubly-fed wind turbine fault monitoring method and system. Background The doubly-fed wind turbine generator has become a main power machine type of global land and offshore wind power by virtue of variable speed constant frequency operation capability, excellent power control performance and higher cost performance, is one of the most widely applied wind power generator types at present, and particularly takes the most role in a large wind power field with the megawatt level or more, and a gear box is used as the most core and vulnerable part in the wind turbine generator, and the operation state of the doubly-fed wind turbine generator directly influences the safety and operation and maintenance cost of the whole machine. In the prior art, the fault monitoring of the gearbox generally adopts a method of directly extracting characteristics based on full-time vibration data, or relies on traditional decomposition methods such as EMD (empirical mode decomposition), wavelet transformation and the like to extract characteristic frequency components related to faults from the vibration data, the vibration data are often mixed in different working conditions, the direct full-time analysis or experience segmentation is difficult to truly ensure the consistency of the working conditions of the data, the same vibration characteristic is extremely easy to cause criterion failure under different working conditions, false alarm is caused, and the problem of low fault monitoring accuracy is easy to occur due to the fact that the traditional decomposition methods such as EMD and wavelet analysis signals are easy to be subjected to modal aliasing. Therefore, it is needed to design a frequency monitoring-based doubly-fed wind turbine generator fault monitoring method and system. Disclosure of Invention Based on the above technical problems, it is necessary to provide a doubly-fed wind turbine generator fault monitoring method and system based on frequency monitoring, which can solve the problem of false alarm caused by the fact that the same vibration characteristic fails under different working conditions in the prior art and the problem of low fault monitoring accuracy caused by the fact that the traditional signal decomposition method is easily subjected to modal aliasing, so as to extract only vibration data of a period corresponding to similar working conditions, eliminate interference of signal difference on the vibration characteristic under different working conditions, and realize early warning of gearbox fault risk as accurately as possible by means of comprehensive criteria based on unit fault by fusion modal energy values. The technical scheme of the invention is as follows: A frequency monitoring-based doubly-fed wind turbine generator fault monitoring method comprises the following steps: Acquiring operation parameter data of the doubly-fed wind turbine, and generating a similar working condition time period of a gear box in the doubly-fed wind turbine according to the operation parameter data based on a clustering algorithm; Extracting target vibration data from the vibration monitoring data of the gear box according to the similar working condition time period; Generating a modal component and a center frequency corresponding to each modal component according to the target vibration data, and generating modal component energy according to the modal component and the center frequency corresponding to each modal component; And generating a fusion modal energy value according to the modal component energy, and generating a unit fault early warning result according to the fusion modal energy value. Optionally, obtaining operation parameter data of the doubly-fed wind turbine, generating a similar working condition time period of a gearbox in the doubly-fed wind turbine according to the operation parameter data based on a clustering algorithm, including: Acquiring operation parameter data of the doubly-fed wind turbine generator, and generating a multidimensional working condition feature vector according to the operation parameter data; normalizing and distributing weight to the multidimensional working condition feature vector, and generating a weighted feature vector; segmenting the weighted feature vector according to a preset segmentation interval, and screening out a target working condition segment based on a clustering algorithm and preset screening conditions; And generating a similar working condition time period according to the target working condition time period. Optionally, the preset segmented interval comprises a wind speed interval and a pitch angle interval; Segmenting the weighted feature vector according to a preset segmentation interval, and screening