Observer Based Fault Estimation for Induction Motor with Stator Inter-turn Short Circuit Faults and Disturbances

The paper studies the fault estimation problem for induction motor with stator inter-turn short circuit faults and disturbances. A T-S fuzzy model is established according to the nonlinear model of a induction motor with stator inter-turn short circuit faults and disturbances. A sliding mode observer is proposed based on the T-S fuzzy model and it is proved that the state estimation error can converge to a certain range. Then condition about the convergence of the output estimation error is analyzed. Further, the injecting equivalent output signal of the sliding mode observer is used to reconstruct the fault signal. By using the H-infinity concept, the conditions are proposed to minimize the effects of disturbances on the fault reconstruction. The constraint conditions are transformed into LMI (Linear Matrix Inequality) to obtain the gains of observer. Finally, the simulation on the induction motor of CRH2 (China Railway High-Speed II) traction system verifies the accuracy and effectiveness of this fault estimation method.