Research On Fault Early Warning Of Main Drive Chain Of Wind Turbine Based On Multi-source Data Fusion

In the past ten years,China’s wind power industry has developed rapidly,and the scale of wind power installed capacity has expanded rapidly.Since wind turbines are mostly located in remote areas and mountainous areas and the operating environment is harsh,the operation and maintenance of wind turbines is a difficult problem.In recent years,more and more wind turbines have provided quality assurance,and more and more attention has been paid to the intelligent operation and maintenance of wind turbines.In the intelligent operation and maintenance of wind turbines,the operation and maintenance of the main drive chain of wind turbines is the most core task.The reason is that the main drive chain of wind turbines is a mechanical component with frequent failures,the longest failure recovery time,and the largest failure loss.Its failure frequency directly affects the reliability and economic benefits of wind farm operation.Therefore,studying the intelligent fault early warning technology of the main drive chain of wind turbines,and timely maintenance of the wind turbine main drive chain before failure occurs,reducing the loss caused by the failure of the main drive chain of wind turbines,and contributing to the healthy development of my country’s wind power industry It is of great significance.This paper takes the main drive chain of doubly-fed wind turbines as the research object,studies the intelligent fault warning technology of the main drive chain of wind turbines,and designs and develops a fault warning system for the main drive chain of wind turbines to realize the main drive chain of wind turbines Smart operation and maintenance.The main research work of this paper includes:1.research on the basic principles of wind turbine main drive chain fault early warning,and on the basis of analyzing the formation mechanism and causes of typical faults in the main drive chain of wind turbines,summarize the feature extraction technology and fault early warning of typical faults in the main drive chain of wind turbines algorithm.This also lays a theoretical foundation for further research on the fault warning technology of the main drive chain of wind turbines.2.considering the limitations of the fault warning technology based on the data source of the wind turbine SCADA monitoring system,a multi-source data fusion method is proposed: fusing the two types of data: the wind turbine SCADA monitoring data and the high frequency vibration monitoring data of the wind turbine main drive chain.Feature extraction,and propose a wind turbine main drive chain fault early warning model based on multi-source data fusion,build a multi-source data fusion fault early warning model based on deep self-encoding network,and judge based on the reconstruction error of the fault features of the multi-type data fusion Whether the main drive chain is operating normally,and whether there is a trend of failure.3.Based on the multi-source data fusion fault early warning model of the deep self-encoding network,the development and engineering application of the fault early warning system of the main drive chain of the wind turbine generator includes: designing the logical architecture,hardware architecture and functional architecture of the fault early warning system,then apply the developed fault warning system to actual wind farms.The results of the trial operation of the fault early warning system show that the researched fault early warning technology can sensitively find the potential defects of the main drive chain,and is of great value for detecting the early faults of the main drive chain of the wind turbine.