System Modelling Approach for Fault Detection in Nonlinear Systems

摘要：

This paper addresses the issue of system modelling for fault detection in nonlinear systems. In many practical situations a primary model (physical model or linearized model) of a nonlinear system already exists. In such cases, we propose to build an auxiliary model that drives the residual of the combined (primary plus auxiliary) model to zero during fault-free operation. The auxiliary model can be built by any data-driven technique in real time. Once the auxiliary model is built and the combined residual converges to zero, the model parameters are kept constant; after this, the model can be used as the basis for fault detection in the original nonlinear system. Simulation shows that the proposed scheme is effective and has potential application ability in fault detection and identification (FDI).

关键词： fault detection:nonlinear system:observer:state estimation:neural networks

作者: Huajun Gong F. N. Chowdhury Jianfang Xing

作者单位: College of Automation Engineering,Nanjing University of Aeronautics & Astronautics,Nanjing,Jiangsu 2 Department of Electrical and Computer Engineering,University of Louisiana at Lafayette,LA 70504,USA

会议类型: 国际会议

会议名称: The 3rd International Symposium on Systems and Control in Aeronautics and Astronautics(第三届航空航天系统与控制国际会议 ISSCAA 2010)

会议地点: 哈尔滨

会议语种:英文

页码: 642-646

在线出版日期: 2010-01-08（万方平台首次上网日期，不代表论文的发表时间）