Research On Intellegent Fault Detection And Fault-tolerant Conrol Of Aeroengine

Failure diagnosis and fault-tolerant control of aero-engine control system is studied for the aero-engine sensors, actuors and components failture, and the problems of state variable modeling, intelligent fault diagnosis, and fault-tolerant controller design are dealt with in this paper.The method for establishing state variable model (SVM) is studied based on the aero-engine component lever model. A large-scale state variable model is gained through combining state variable models. The comparison of Kalman filter and T-S (Takagi-Sugeno) fuzzy system Kalman filter approaches for aero-engine fault estimation is illustrated. An aero-engine sensor fault diagnostics system is proposed by a bank of Kalman filters and the weighted sum of squared residual, which has a better performance on sensor fault detection and accommodation. The problem of robust fault detection filter is investigated for an aero-engine system with unknown model error and fault inputs. Attention is focused on formulating the design of the residuals match with reference model. The robustness of the residual system is proved to be stable. The existence condition of the above filters, which can guarantee the robustness against disturbances and sensitivity against fault are solved by LMI. And an aero-engine component neural network fault detection approach is also proposed in this paper.An adaptive neural network control designed approach is proposed for nonlinear system which contains nonlinear model error and fault, based on backstepping technology, then using the method to estimate the on-line term of model error and fault. The robust term is utilized to compensate approximation model error. When the fault term has being detected, it is designed to revise virtual controller to compensate the fault term. The effect of fault is eliminated by the fault-tolerant controller. A robust fault-tolerant control approach is proposed for aero-engine system with some uncertain faults. The T-S (Takagi-Sugeno) fuzzy systems are employed to represent the aero-engine nonlinear system. The robust output feedback controller can satisfy the H∞performance index and the requirement of close-loop system stability.