Multi-wind turbine generator operation state identification method based on migration component analysis

Abstract

The invention discloses a multi-wind turbine generator operation state identification method based on migration component analysis, and belongs to the technical field of monitoring and analysis of the running states of the wind turbine generators, health condition evaluation and power generation performance evaluation. Firstly, key influence parameters of wind power are mined, and a wind turbine generator operation characteristic data set is constructed; then, a multi-wind turbine generator operation data distribution assimilation model based on migration component analysis is constructed, and a multi-wind turbine generator operation data distribution assimilation data set is obtained; then, a target wind turbine generator normal behavior model based on machine learning is constructed; and finally, on the basis of the normal behavior model of the target wind turbine generator, power prediction residual errors of different wind turbine generators are obtained, and batch identification of the running states of the multiple wind turbine generators is realized. The model can give consideration to the identification precision and efficiency at the same time, and can provide reliable data support for the monitoring and analysis of the operation state of the wind turbine generator, the evaluation of the health condition of the wind turbine generator and the evaluation of the power generation performance of the wind turbine generator.

Description

technical field [0001] The invention relates to the technical fields of wind turbine operating state monitoring and analysis, health status assessment, and power generation performance evaluation, and in particular to a method for identifying operating states of multiple wind turbines based on migration component analysis. Background technique [0002] The working environment of wind turbines is harsh, and the operating conditions are complex and changeable. The superposition of various factors may cause the wind turbines to deviate from the normal operating state, resulting in a decline in the health status and power generation performance of the wind turbines. Accurate identification of the operating state of wind turbines can realize the effective evaluation of the health status and power generation performance of wind turbines, and is of great significance to improve the safety, reliability and economy of wind turbines. Existing methods usually use data from supervisory ...

AI Technical Summary

Problems solved by technology

However, due to differences in geographic location, wind resource endowment, and unit health status, the SCADA data distribution of different wind turbines is significantly different. If the trained normal behavior model of a single wind turbine is directly applied to the identification of the operating status of multiple wind turbines, the identification accuracy Low; in order to improve the identification accuracy, it is necessary to carry out repetitive training for the normal behavior model of each wind turbine, which requires a large workload and takes a long time

Therefore, the existing methods cannot be applied to the identification of the operating status of multiple wind turbines, and there is a problem that the identification accuracy and efficiency cannot be considered at the same time.

Images