CN-121996988-A - Wind turbine generator system fault detection method and device and computer device

CN121996988ACN 121996988 ACN121996988 ACN 121996988ACN-121996988-A

Abstract

The disclosure relates to a wind turbine generator system fault detection method, a wind turbine generator system fault detection device and a computer device. The wind power failure monitoring method and device relate to the field of wind power failure monitoring and solve the problem of low running efficiency of the unit caused by failure detection and response lag. The method comprises the steps of obtaining a fault data set and a normal data set according to historical operation data of a wind turbine generator, wherein the fault data set comprises operation data related to faults before the faults occur, the normal data set comprises operation data in a normal working state, obtaining at least one difference characteristic between the fault data set and the normal data set, and generating a fault early warning rule according to the at least one difference characteristic. The technical scheme provided by the disclosure is suitable for management of the wind generating set, realizes early warning of faults such as vibration and the like, and provides support for timely converting operation postures so as to avoid serious faults from influencing operation efficiency.

Inventors

JI HAIPENG
XIA DEXI
Wen yaoguang
WANG ZHENGJUN
YUAN FEI
ZHANG YICHENG

Assignees

国电联合动力技术有限公司

Dates

Publication Date: 20260508
Application Date: 20260407

Claims (10)

1. The wind turbine generator system fault detection method is characterized by comprising the following steps of: acquiring a fault data set and a normal data set according to historical operation data of the wind turbine generator, wherein the fault data set comprises operation data related to faults before the faults occur, and the normal data set comprises operation data in a normal working state; Acquiring at least one difference characteristic between the fault data set and the normal data set; and generating a fault early warning rule according to at least one difference characteristic.
2. The wind turbine fault detection method of claim 1, wherein the historical operating data includes at least any one or more of the following parameters of the wind turbine: wind speed, opposite wind angle, impeller rotating speed, pitch angle pitch speed, power, cabin front-back vibration and cabin left-right vibration, The step of obtaining the fault data set and the normal data set according to the historical operation data of the wind turbine generator comprises the following steps: Determining at least one fault time point according to a preset fault detection condition; And constructing the fault data set according to the historical operation data in the first time interval before each fault time point, and constructing the normal data set according to the historical operation data outside the fault data set.
3. A wind turbine fault detection method according to claim 2, wherein the step of obtaining at least one difference feature between the fault dataset and the normal dataset comprises: Acquiring a fault value of each characteristic index in the fault data set and a normal value in the normal data set, wherein the fault value is contained in a preset characteristic index list, and the characteristic index list contains at least one characteristic index; And taking the characteristic index of which the difference degree between the fault value and the normal value meets a preset distinguishing degree condition as the difference characteristic.
4. The wind turbine generator system fault detection method according to claim 3, wherein the step of obtaining the fault value of each of the feature indexes included in the preset feature index list in the fault data set and the normal value in the normal data set includes: Sliding the normal data set into a plurality of first sliding windows and sliding the fault data set into a plurality of second sliding windows according to the first time interval; extracting the characteristics of the first sliding windows one by one according to the characteristic index list, obtaining first characteristic data points corresponding to the first sliding windows, and constructing a normal characteristic data set based on all the first characteristic data points; Feature extraction is carried out on the second sliding windows one by one according to the feature index list, second feature data points corresponding to the second sliding windows are obtained, and a fault feature data set is built based on all the second feature data points; And carrying out statistical analysis on the normal characteristic data set and the fault characteristic data set to obtain the normal value and the fault value of each characteristic index in the characteristic index list.
5. The wind turbine generator system fault detection method according to claim 3, wherein the step of using the characteristic index, in which the degree of difference between the fault value and the normal value meets a preset degree of distinction condition, as the difference characteristic includes: calculating the effect of the fault value and the normal value of each characteristic index; and taking the characteristic index of which the effect magnitude value meets the preset distinguishing degree condition as the difference characteristic.
6. A wind turbine fault detection method according to claim 3, wherein said step of sliding said normal dataset into a plurality of first sliding windows and said fault dataset into a plurality of second sliding windows according to said first time interval is preceded by the step of: and acquiring a plurality of characteristic index lists, wherein different characteristic index lists are associated with different fault types to indicate the subsequent characteristic extraction according to the different fault types.
7. The wind turbine fault detection method according to claim 1, wherein the step of generating a fault pre-warning rule according to at least one of the difference features comprises: Performing value analysis on the difference features through a decision tree to obtain optimal fault intervals of the difference features; Constructing an independent basic rule based on the single difference feature and the corresponding optimal fault interval; generating the fault pre-warning rule comprising at least one of the independent basic rules.
8. The wind turbine fault detection method of claim 1, further comprising: and under the condition that an event conforming to the fault early warning rule is detected, generating an application decision, wherein the application decision comprises a control instruction for the wind turbine.
9. The utility model provides a wind turbine generator system fault detection device which characterized in that includes: the data set dividing module is used for acquiring a fault data set and a normal data set according to historical operation data of the wind turbine generator, wherein the fault data set comprises operation data related to faults before the faults occur, and the normal data set comprises operation data in a normal working state; The feature screening module is used for acquiring at least one difference feature between the fault data set and the normal data set; And the fault rule generation module is used for generating a fault early warning rule according to at least one difference characteristic.
10. A computer apparatus, comprising: A processor; A memory for storing processor-executable instructions; Wherein the processor is configured to perform the wind turbine fault detection method of any one of claims 1 to 8.

Description

Wind turbine generator system fault detection method and device and computer device Technical Field The disclosure relates to the field of wind power failure monitoring, in particular to a wind turbine generator failure detection method, a wind turbine generator failure detection device and a computer device. Background In recent years, the wind power industry is developed at a high speed, the construction scale of a wind farm is continuously enlarged, and the project development period is continuously shortened. Under the pressure of cost control and construction period, the problems of safety coefficient reduction, insufficient control of manufacturing and assembly quality and the like occur in part of wind turbine generators in the design stage, and hidden danger is generated for long-term stable operation of the wind turbine generators. Meanwhile, the wind turbine generator system is iterated rapidly towards the large-scale direction, the diameter of the impeller and the wind sweeping area are continuously increased, and the pneumatic load and the mechanical load born by the wind turbine generator system are unbalanced. Under the severe wind resource conditions of complex wind conditions and rapid and frequent wind shear, the traditional unit control strategy is difficult to rapidly adapt to the dynamically-changed external working conditions, and the problems of control response lag, unbalanced load distribution and the like are easy to occur, so that the typical faults of abnormal vibration of a unit cabin, aggravation of driving chain impact and the like are caused, the unit is forced to operate with limited power or is stopped in an unplanned manner, and the power generation efficiency and economic benefit of a wind power plant are greatly reduced. Disclosure of Invention In order to overcome the problems in the related art, the disclosure provides a method and a device for detecting faults of a wind turbine generator and a computer device. By analyzing the historical operation data, the data characteristics before the occurrence of the faults are obtained, and then the early warning is generated, the faults of the wind turbine generator set can be found in early stage, the problem of low running efficiency of the wind turbine generator set caused by fault detection and response lag is solved, the early warning of faults such as vibration is realized, and support is provided for timely converting the running gesture to avoid serious faults from influencing the running efficiency. According to a first aspect of embodiments of the present disclosure, there is provided a wind turbine generator fault detection method, including: acquiring a fault data set and a normal data set according to historical operation data of the wind turbine generator, wherein the fault data set comprises operation data related to faults before the faults occur, and the normal data set comprises operation data in a normal working state; Acquiring at least one difference characteristic between the fault data set and the normal data set; and generating a fault early warning rule according to at least one difference characteristic. Further, the historical operation data at least comprises any one or more of the following parameters of the wind turbine generator: wind speed, opposite wind angle, impeller rotating speed, pitch angle pitch speed, power, cabin front-back vibration and cabin left-right vibration, The step of obtaining the fault data set and the normal data set according to the historical operation data of the wind turbine generator comprises the following steps: Determining at least one fault time point according to a preset fault detection condition; And constructing the fault data set according to the historical operation data in the first time interval before each fault time point, and constructing the normal data set according to the historical operation data outside the fault data set. Further, the step of obtaining at least one difference feature between the faulty data set and the normal data set includes: Acquiring a fault value of each characteristic index in the fault data set and a normal value in the normal data set, wherein the fault value is contained in a preset characteristic index list, and the characteristic index list contains at least one characteristic index; And taking the characteristic index of which the difference degree between the fault value and the normal value meets a preset distinguishing degree condition as the difference characteristic. Further, the step of obtaining the fault value of each feature index in the fault data set and the normal value in the normal data set, where the fault value is included in the preset feature index list, includes: Sliding the normal data set into a plurality of first sliding windows and sliding the fault data set into a plurality of second sliding windows according to the first time interval; extracting the characteristics of the first sliding windows o