Fault-detection filter for networked control systems with packed dropouts and access constraints

This paper investigates the problem of fault detection for networked control systems with packet dropout and access constraints. The packed dropout, which is modeled as a Bernoulli distribution, exists in the channels from controllers to actuators. The access constraints exist in the channels from sensors to controllers and controllers to actuators. By using linear matrix inequality, a fault detection filter is designed. Then, a new threshold calculation method is proposed, so that residual evaluation function can quickly exceed the designed threshold and the residual error can achieve minimum when actuator occur fault. Finally, simulation results show that the proposed method in this paper can detect fault faster.