Migration diagnosis method of the gearbox fault of a wind turbine generator system

Abstract

The invention belongs to the technical field of condition monitoring and fault diagnosis of a wind turbine generator system, in particular to a migration diagnosis method of the gearbox fault of a wind turbine generator system. The method comprises the following steps: establishing four neural network structures, namely, a source domain feature extractor, a target domain feature extractor, a domain classifier and a domain discriminator; obtaining predictive label values by forward propagation from annotated source domain data, network training loss functions are calculated according to predictive label and actual label, and source domain feature extractor and domain classifier are pre-trained by back propagation algorithm. The loss functions of the source domain feature, the target domainfeature and the domain discriminator are calculated by forward propagation from the source domain data and the target domain data, and the domain discriminator and the target domain feature extractorare trained by back propagation algorithm respectively. The newly acquired target domain data is input into the target domain feature extractor, the feature is calculated, and the predictive label ofthe new data is obtained by the domain classifier input from the feature.

Description

technical field [0001] The invention belongs to the technical field of state monitoring and fault diagnosis of wind turbines, and in particular relates to a fault migration diagnosis method of a gearbox of a wind turbine. Background technique [0002] Wind turbines operate in harsh outdoor natural environments for a long time, and the failure rate is higher than that of conventional generators. At the same time, the wind turbine is located in a remote location, and the nacelle is installed at a height of nearly 100 meters, which causes great inconvenience to the maintenance and repair of the wind turbine, and is an important reason for its high maintenance cost. According to incomplete statistics, based on the 20-year life cycle operating cost of wind turbines, the downtime caused by failures has accounted for 25.6% of the rated power generation time, and the maintenance costs have reached 20-25% of the total income of wind farms. [0003] The gearbox is the key equipment c...

AI Technical Summary

Problems solved by technology

[0007] Although some achievements have been made in wind turbine gearbox fault diagnosis, most of the methods still need to rely on expert experience, which is subjective and difficult to describe clearly in a formal way. In wind farms with different models and working conditions, these methods will not be universal and popular

In addition, the fault data of large components of wind turbines is very scarce, especially the data with fault labels is very scarce, which makes the state recognition algorithm based on supervised learning unable to be really applied due to the lack of fault samples

In addition, as the gear box of the unit ages, its performance gradually decreases, and the distribution of the signal representing the state of the gearbox will also change accordingly. If the model does not have the ability of adaptive adjustment, this will seriously affect the generalization of the diagnostic model ability

Images