Working Condition Identification And Multi-Condition Dynamic Behavior Analysis Of Wind Turbine Main Drivetrain

Benefiting from technological advances and business model innovations,the wind energy industry is growing rapidly.However,with the continuous increase in the installed capacity of wind turbines and the aging of early service wind turbines,the operation and maintenance costs have been continuously increasing.The main drivetrain of wind turbines has a high frequency of abnormal failures due to the complex and diverse operating conditions and unclear multi-working condition dynamic behavior,greatly affecting the economy and safety of wind turbines.Therefore,it is important to identify the working conditions of the main drivetrain and analyze its multi-working condition dynamics.Aiming at the problems that the current working condition identification method is affected by the wind turbine anomalies,and the multi-working conditions dynamics behavior of the main drivetrain is unclear,the main drivetrain of double-fed wind turbine is selected as the research object.Based on the data collected by the Supervisory Control and Data Acquisition(SCADA),the working condition identification method of the wind turbine main drivetrain is optimized,solved the problem that the identification results are affected by the main drivetrain operation state due to the improper selection of input parameters.The study on the multi-working conditions dynamics of the main drivetrain is carried out,revealed the multi-working condition dynamic behavior of the main drivetrain,and clarified the tooth root crack fault sensitive working conditions and the variations of the fault characteristic indexes in each working condition.It provides conditions for wind turbine main drivetrain condition monitoring,condition evaluation,fault diagnosis and so on.The main research work includes:(1)The wind turbine working principle,the main drivetrain basic structure and the SCADA system are outlined,and the main parameters of the research object are presented.The SCADA historical data is cleaned to remove the abnormal value and normalized to eliminate the effect of dimension.Parameter support and data preparation are provided for the subsequent work.(2)A wind turbine main drivetrain torsional dynamics model,which considering the time-varying meshing stiffness,damping,tooth backlash and other nonlinear factors,is established and solved,and the accuracy of the model is verified by wind farm field tests.By using the improved potential energy method,a time-varying meshing stiffness model for normal and crack fault states is developed,and the effect of root crack expansion on the meshing stiffness is analyzed.Model basis is provided for the subsequent research on the analysis of the wind turbine main drivetrain multi-working condition dynamics.(3)The wind turbine main drivetrain working condition identification optimization method is proposed,and solved the problem that the identification results are affected by the main drivetrain operation state due to the improper selection of input parameters.In this method,after selecting the feature parameters and classifying the main drivetrain historical working conditions,a working condition identification model is established,which uses the main drivetrain inputs as the identification parameters,and verified by cases.The method can ensure the accuracy of working condition identification while addressing the problem that while the main drivetrain output is used as the working condition identification parameter,the result of working condition identification will be affected by the wind turbine anomalies and lead to misjudgment of working condition recognition,and provide conditions for main drive system condition monitoring and condition evaluation.(4)The research on the dynamic characteristics of wind turbine main drivetrain in normal and fault state is carried out,and clarified the crack fault sensitive working conditions and the variations of the fault characteristic indexes in each working condition.The dynamic characteristics of the main drivetrain each gear pair,the feature of tooth root crack fault and the dynamic characteristics of different fault levels are studied.The multi-working condition dynamics characteristics in the main drivetrain normal and faulty states are analyzed,revealed the multi-working condition dynamic behavior,and clarified the tooth root crack fault sensitive working conditions and the variations of the fault characteristic indexes in each working condition.The method provides an important theoretical basis for the fault diagnosis of the main drivetrain.