Multi-Agent Based Fault Detection and Isolation in More Electric Aircraft

In this paper, a multi-agent system is developed to detect mechanical faults occurring in the flight control system in a more electric aircraft. The studied system is composed of electrical actuators controlling ailerons. In such an interconnected system, a distributed fault detection and isolation (FDI) is required to improve the reliability of the fault detection. Here, a multi-agent system is employed for this purpose. Each agent includes an FDI filter providing residual signals for its own actuator and its neighbors. A fault is detected and isolated when a majority of agents vote for it. Simulation results show the validity of this approach. Experimental tests are in progress to confirm the effectiveness of the proposed approach.