基于信息熵的纵列式无人直升机故障树诊断  

Information Entropy-based Fault Tree Analysis of Unmanned Tandem Helicopter

作　　者：刘世前[1] 施维[1] 沈勇 朱亮[1] 

机构地区：[1]上海交通大学航空航天学院,上海200240 [2]上海航空测控技术研究所,上海200230

出　　处：《系统仿真学报》2009年第14期4424-4428,共5页Journal of System Simulation

基　　金：国家航空联合基金(10577012;60705006)

摘　　要：研究纵列式无人直升机故障诊断问题。首先根据其操纵控制系统的结构特征,建立操纵控制故障树及故障传播矩阵。其次运用信息熵决策推理方法,分析样本测量与信息不足引起的不确定决策,进行故障树不确定性推理,并确定故障节点重要度及隶属关系。最后根据二元假设检验,进行故障判定。并针对直升机尾传动轴不对中故障,引起振动与作动器效能降低,进行仿真与实验,验证该故障诊断方法的可行性与有效性。The problem of fault diagnosis of an unmanned tandem helicopter was studied. Firstly, Fault Tree （FT） and its transferring function matrix was established according to the structural feature of the helicopter flight control system. Secondly, an information entropy based reasoning method was employed to the uncertainty fault reason for insufficient sample data and information, and to determinate importance degree of the probability of the FT nodes and their membership relation. Finally, node fault of the FT was judged by using two-element hypothesis test. The availability and effectiveness of the proposed fault method was verified by a simulation and test example, where a fault of the rear-axis of the helicopter was unsymmetrical such that mechanical vibration was produced and the control effect of the actuator in the rear rotor was attenuated.

关 键 词：纵列式无人直升机 信息熵 故障树 不确定性推理 飞行控制 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] V279[自动化与计算机技术—控制科学与工程]