Research Of Sensor Fault Diagnosis Based On Model

At present the reliability of the control system was paid great attention. In order to improve the reliability and security of the system, detect the breakdowns in real time, analysis the cause and the features of it, so as to prevent the trouble occurred, based on the current situation, make the research of the sensor fault diagnosis based on the model, and the work mainly been engaged in is as follows:In the beginning, combined with years of the fault diagnosis technique development and the present situation, proposed the key work is studied in this paper. Then it introduces various fault diagnosis method based on the sensor fault diagnosis, the principles and procedures, and discusses two kinds of methods to improve performance of fault diagnosis, unknown input observer and threshold method.To carry out sensor fault diagnosis fatherly, penetrate a method that based on unknown input observer sensor fault diagnosis by equivalence transformation. By a series of changes, eliminate the interference with never losing any information of the fault. It is able to detect the systems fault in the case that no unknown input existing. Simulations validate the method for separating and identifying fault is effective.Coming up with a fault diagnose method that based on strong tracking filter for nonlinear systems, and establishing the model for three kinds of sensors fault respectively. To discuss the specific approach the related fault diagnosis of strong tracking filters and the application in a constant sensor fault. Simulations verified that the proposed method of sensors fault diagnosis for nonlinear system is effective.Finally, as a summary, a brief conclusion of this text is given and further research in this field is also discussed.