Fault-tolerant control of aero-engine based on backstepping

摘要：

An adaptive neural network backstepping control approach is proposed for aero-engine system with faults. A method was employed to estimate the terms of model error and fault on-line. The proposed controller is utilized to compensate the model error approximately. When a fault is detected, controller is reconstructed to accommodate the fault. Therefore, the fault influence is eliminated by the fault-tolerant controller. It is proved that the residual signal converges to an arbitrary small neighborhood around zero. Finally, the low pressure turbine speed controller is designed based on a nonlinear model of a turbofan engine. Simulation results show the feasibility of the proposed approach.

关键词： aero-engine backstepping fault-tolerant neural network nonlinear system

作者: Zi-jie Zhu Xiang-hua Huang

作者单位: College of Energy and Power Engineering, Nanjing University of Aeronautics and Astronautics Nanjing, College of Energy and Power Engineering,Nanjing University of Aeronautics and Astronautics Nanjing,

会议类型: 国际会议

会议名称: The 3rd International Symposium on Jet Propulsion and Power Engineering(第三届喷气推进与动力工程国际会议 ISJPPE)

会议地点: 南京

会议语种:英文

页码: 579-582

在线出版日期: 2010-09-13（万方平台首次上网日期，不代表论文的发表时间）