Multiple fault parameter identification based on extremum seeking for multi-agent systems

Abstract

In this paper, a multiple fault parameter identification method based on extremum seeking algorithm is proposed for the consensus of multi-agent systems. Firstly, a model for the cooperative consensus multi-agent systems is established based on the graph theory and linear system theory. Secondly, by using the proposed cooperative consensus model, an adaptive multiple fault parameter identification method based on extremum seeking algorithm is designed. Then, the multiple fault parameter identification problem is reformulated as an optimal input problem of multiple extremum seeking. Thirdly, the multiple fault closed-loop search framework based on gradient information is given and the extremum seeking cost function is designed. Finally, the performance of the multiple fault parameter identification algorithm using extremum seeking is rigorously evaluated through a series of simulations. In addition, the extremum searching performance under different excitation signals is discussed.