Optimal diagnostic observer for sampled-data systems
采样数据系统最优诊断观测器设计

The direct design method of optimal diagnostic observer for sampled-data systems is investigated. Firstly, the Luenberger condition of the diagnostic observer for the general discrete-time systems is studied from the transfer function viewpoint. By this condition, an appropriate diagnostic observer for sampled-data systems is built and its hybrid dynamics is derived. Then, the sampled-data fault detection problem is formulated as a ratio-type optimization problem. By using the co-inner-outer factorization technique and the algebra method, we obtain a general optimal solution which makes the discrete-time residuals robust to the continuous-time unknown inputs, but sensitive to faults. Simulation results of a numerical example are presented to demonstrate the effectiveness of the proposed method.