未知离散故障下切换电路系统的故障参数辨识  

Fault parameter identification for switched circuit system under unknown discrete faults

作　　者：肖晨雨 郁明 Xiao Chenyu;Yu Ming(School of Electrical and Automation Engineering, Hefei University of Technology, Hefei 230041, China)

机构地区：[1]合肥工业大学电气与自动化工程学院

出　　处：《仪器仪表学报》2018年第4期195-204,共10页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61673154)项目资助

摘　　要：研究了切换模拟电路系统在未知离散故障条件下的故障参数估计方法。首先,利用混杂键合图（HBG）理论对切换电路系统进行建模,通过因果路径覆盖法推导出系统的全局解析冗余关系（GARRs）用以检测故障,并根据GARRs构造故障特征矩阵（FSM）用以隔离可能故障参数集合。为了修正故障集合,提出一种新的优化算法——基于微调机制的生物地理学优化算法（PABBO）,将故障隔离获得的可能故障参数集合和未知离散故障的模式变化时间视为优化算法的设计变量,实现故障参数和未知离散故障的同时辨识。通过仿真与实际实验结果分析,验证了基于HBG和GARRs的混杂系统故障检测方法的有效性,以及基于PABBO的故障参数和模式变化时间同时辨识算法的可行性。In this paper,the estimation method of fault parameters in switched analog circuit system under unknown discrete fault condition is investigated. Firstly,hybrid bond graph（ HBG） theory is used to model the switched circuit system,the global analytical redundancy relationships（ GARRs） of the system is derived via causal path coverage method to detect faults,and the fault signature matrix（ FSM） is constructed according to GARRs to isolate the possible fault parameter set. In order to correct the faults set,a new optimization algorithm called biogeography-based optimization algorithm with pitch adjustment（ PABBO） is proposed,where the possible fault parameter set obtained with fault isolation and the mode switch time of unknown discrete fault are considered as the design variables of optimization algorithm,the simultaneous identification of fault parameters and mode switch time of unknown discrete fault. Through the analysis of simulation and actual experiment results,the effectiveness of HBG and GARRs based fault detection method of hybrid system is verified,and the feasibility of simultaneous identification of fault parameters and mode switch time based on PABBO is proved.

关 键 词：混杂键合图 全局解析冗余关系 故障特征矩阵 基于微调机制的生物地理学优化算法 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置] TH851[自动化与计算机技术—控制科学与工程]