Decentralized Fuzzy Polynomial Observer Synthesis for Large-Scale Nonlinear System With Arbitrary Faults and Uncertainties

This paper proposes a method to synthesize a decentralized polynomial observer for the large-scale nonlinear system. The polynomial Takagi-Sugeno (T-S) fuzzy framework is employed to model the large-scale nonlinear system to increase the accuracy and decrease the complexity with respect to applying T-S model or linearization method in previous studies. The faults and uncertainties are taken into consideration in this paper. The faults and uncertainties in this work are arbitrary and do not need to satisfy the bounded constraints that will make the observer synthesis for large-scale nonlinear systems much more challenging. Additionally, the large-scale polynomial T-S fuzzy system will be modified to a new framework before starting the observer synthesis. The method for designing the decentralized fuzzy polynomial observer is proposed in this paper. The decentralized fuzzy polynomial observer not only completely eliminates the impact of uncertainties and faults but also estimates unknown states and reconstructs the faults of large-scale nonlinear systems. The sufficient conditions for designing the decentralized polynomial observer will be presented in the theorems. Finally, an illustrative example is provided to show the superiority and success of the proposed method.