Research On Fault Diagnosis Of Megawatt Wind Energy Conversion System

The high failure rate of the Wind Energy Conversion System(WECS)results in expensive maintenance.Therefore,how to detect and isolate faults under the characteristics of random wind is the focus of this research.A new WECS integrated fault detection and isolation(FDI)algorithm is proposed,which can accurately estimate the continuous state and discrete state of the wind turbine with unknown fault inputs.The principle is to construct a system fault residual based on the proposed accurate estimate of the estimator,and use statistical decision tools to achieve fault detection and isolation.The continuous state estimation of the system can be obtained by the robust hybrid estimator,and the discrete state estimation of the system can be derived by the Bayesian theory.Firstly,the stochastic linearhybrid model of Wind Energy Conversion System is studied in this paper,wh ich can reflect the normal and fault characteristics of wind turbines and introduce Markov transition matrix to accurately describe the stochastic dynamic characteristics of wind turbines.Secondly,a robust hybrid estimator is designed to estimate the state of the WECS based on the proposed model.Compared with the existing wind energy conversion system estimator,the proposed robust hybrid estimator is more consistent with the characteristics of the WECS due to its discrete working state.Finally,the FDI algorithm is researched to detect and isolate faults in the WECS through using the robust hybrid estimator.It can be analyzed that the stochastic model predictive control(SMPC)has a precise control performance and the FDI algorithm possesses well effect on fault detection and isolation for the WECS through the simulation analysis.The proposed method can effectively estimate the fault magnitude and fault occurrence.