Multiple Anfis Oservers Based Sensor Fault Detection and Control in a Satellite Launcher

Sensor failure detection and identification has been considered as an important issue, particularly when the measurements from sensors are used in the feedback loop of a control law. Kalman filters and Luenberger observers have been widely used to generate signal redundancy by means of state estimation. Dedicated observer scheme and generalized observer scheme are the older methods available for the evaluation of the residual to distinguish a particular fault from other. These methods are based on quantitative models of the system dynamics and these applications are limited to linear systems. To overcome the problems due to modeling error and non-linearity of the system, the proposed intelligent FDI system uses Multiple Adaptive Neuro-Fuzzy Inference System (MANFIS) which is functionally equivalent to Sugeno fuzzy model. ANFIS is the combination of an ANN and FIS in which observers are developed as FIS and ANN is used to determine the parameters of this fuzzy system. Since these observers are designed based on input-output relationships instead of mathematical model of a process, this FDI overcomes the problem of modeling errors and the difficulties encountered in developing accurate analytical observers using mathematical model. In this work, such fault detector is designed and simulated for a satellite launcher. The individual failures of three sensors in the satellite launcher are considered and the results are discussed. The results show that the system is able to detect any sensor failure situations perfectly.