Research on State Detection Method of Complex System Based on Unknown Input

In order to improve the abnormal detection and fault diagnosis ability of the complex system itself, a complex system state evaluation method based on unknown input was proposed by means of parameter disturbance robustness analysis. In this method, the baseline of characteristic parameters is described dynamically and the error between the measured output and the estimated output is used to construct the dynamic threshold. Finally, taking a complex electromechanical system as the application object, the proposed method is verified and analyzed based on the actual running data. The results show that the proposed method takes the uncertainty of parameters into account, and can effectively detect the abnormal state of the system, reducing the false alarm and missing detection.