State and faults estimation via Artificial Neural Networks

Abstract

This paper deals with the state and fault estimation for non linear systems modeled using the Takagi Sugeno approach. An artificial neural network with unknown inputs is used in the objective of estimate state and faults affecting the system. Firstly, the problem of state estimation is considered. In second step, the proposed approach is extended to the actuator fault estimations. The proposed method is applied to an academic example to show its efficiency.