An Improved Method On Estimating The Unknown Input Distribution Matrix

• Main Authors: Wei-Lun Chang, 張瑋倫
• Other Authors: Daw-Shang Hwang
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/68943260917772408607

碩士 === 國立臺灣海洋大學 === 輪機工程系 === 95 === Modern control systems are becoming more and more complex, consequently the role of fault diagnostic systems are becoming important. In all approaches of fault diagnosis systems, the model-based fault diagnosis systems are accepted and used popularly for its cost efficiency and reliability. Since disturbances and modeling error are unavoidable in the system’s model, they will affect the normal operation of fault diagnosis system, therefore, robustness of the fault diagnostic system is an important issue such that the diagnostic results should be immune to the disturbance or modeling error, only related to the present faults. Unknown Input decoupling approach is an important method for designing a robust fault diagnostic system. However, the robust fault diagnostic design required that the disturbance distribution matrix E should be known in a priori. Patton&Chen provide a design procedure for the estimation of the matrix E, but this approach can only be used in the system with the output matrix C is square and invertible. This restriction obstructs the use and development of the estimation technique. In this thesis, we provide a new design procedure to overcome this problem such that unknown input distribution matrix E can be estimated with less sensor. Simulation results demonstrate that the developed approach is practicable and effective.