中文摘要：

针对大型民机飞行控制系统作动器故障,考虑飞机对象和飞行控制系统结构参数随高度和马赫数时变,且测量信号受干扰影响,提出线性参变（linear parameter varying,LPV）系统鲁棒故障检测方法.目标是寻找参变滤波器使残差系统实现参变H∞/H-混合指标,利用回路成形思想将该目标转换为参变H∞鲁棒性能问题,进而使用线性分式变换转换为具有结构不确定性的线性时不变H∞鲁棒性能问题,运用标定有界实引理推导出故障检测滤波器存在的充分条件,并借助Matlab半正定规划工具箱求解滤波器参数,得到的滤波器与被监控对象具有相同的时变参数结构,能够随可测参数自适应调参.最后以B747-100/200大型民机为例,检测其升降舵舵效损伤故障,仿真结果表明本文提出的LPV鲁棒故障检测方法是有效的.

英文摘要：

A robust fault detection filter design method for faults in the actuator of a jumbo jet transport aircraft is proposed by considering how the structural parameters of the transport aircraft plant and its flight control system vary with flight altitude and Mach number, and how the measure signals are disrupted by noise. The ob jective is to find a parameter-vaying filter that makes the residual system fulfill a parameter varying mixed H∞/H index. Using loop-shaping theory, the mixed index is transformed into a parameter-varying H∞ index. The LFT tool is then employed to transform the above problem into a H∞ robust performance problem for an LTI system with structured uncertainty, and the sufficient conditions for the existence of the filter are achieved via scaled bounded real lemmas. The parameters of the LPV filter for fault detection are obtained by the Matlab SDP toolbox. This LPV filter has the same varying parameter structure as the monitored plant, and it has adaptive capability for the varying parameters. Finally, a simulation example of the elevator surface injury of a B747-100/200 jumbo jet is illustrated to demonstrate the practical effect of the proposed approach.