Condition Monitoring Of Wind Turbine Based On SCADA Data

With the rapid development of wind power market,the increase of total installed capacity and the capacity of wind turbine unit and the complexity of the operating environment put forward higher requirements for operation and maintenance ofwind farm.Thus,wind turbine condition monitoring has been a high degree of concern.Current condition monitoring methods for wind turbine are mostly subsystem-level monitoring,which can not consider the interaction of components and subsystems on the overall level,and most monitoring methods need to install additional monitoring and diagnosis equipment.SCADA(Supervisory Control and Data Acquisition)system,which is standard in wind turbines,can monitor all subsystems and components extensively.But its monitoring method is limited to alarm for single parameter value exceeding the limit,which can not prevent the deterioration of equipment failure.The fault alarm results do not take into account the interaction between components and subsystems,which can not provide effective fault diagnosis technical support.Therefore,this paper aims to explore the availability of SCADA data,and study the condition monitoring method of wind turbine based on operational monitoring data.Hope it provide the idea for further study of condition monitoring and fault diagnosis for wind turbine based on operation data.Based on the actual data of the in-service wind turbine,combined with the operation principle,the monitoring physical quantity and the SCADA data reflecting the state of the component subsystem are selected.Least squares method is used to establish the least squares matrix of the normal state of wind turbine to quickly select the units with different operating states.As for the abnormally operating unit,the polynomial coefficient model is established by the polynomial regression method to locate the faulty parts or subsystems in the normal condition.