完备ARRs集的生成方法及传感器优化配置应用  

Method for Generating Complete ARRs Set and Its Application in Sensor Placement Optimization

作　　者：杜敏杰[1] 蔡金燕[1] 刘利民[1] 陈鹏[1] 

机构地区：[1]军械工程学院光学与电子工程系,石家庄050003

出　　处：《数据采集与处理》2012年第5期565-569,共5页Journal of Data Acquisition and Processing

基　　金：河北省重点基础研究(1096529D)资助项目

摘　　要：广泛应用于故障诊断和传感器优化、分析、证实的解析冗余关系(Analytical redundancy relations,ARRs)缺乏系统、有效的方法来产生完备ARRs集,为此,提出了一种逐次消元法。该方法以系统元关系(Primary relations,PRs)为基础,通过若干次循环消元过程,生成了完备ARRs集,同时生成了对应的假定特征矩阵(Hypothetical signature matrix,HSM);基于HSM,把传感器优化配置问题映射为一个特殊的0-1整数规划模型,并用分支定界法求解该模型。应用表明,该方法能在不降低故障检测率、隔离率的前提下减少传感器数目,降低了测试代价,对故障诊断中的传感器配置问题有借鉴意义。Analytical redundancy relations（ARRs） are frequently used in the area of fault diagnosis,as well as optimizing,analyzing,and validating of sensors in a system.However,less attention has been paid to the development of systematic and efficient approaches for generating complete ARRs set.Hereto,an efficient method,named the successive elimination is presented.Based on the primary relations（PRs） of the system,the method generates complete ARRs set and the consequent hypothetical signature matrix（HSM） by several elimination loops.Based on the strength of HSM,the optimal sensor placement problem is mapped onto a special case of 0-1 integer programming（IP） problem.Then,the problem is solved by an algorithm of branch-and-bound.Application results show that the method can decrease sensor numbers and testing cost without decreasing fault detection rate（FDR） and fault isolation rate（FIR）.And it is beneficial to the sensor placement problem for fault diagnosis.

关 键 词：故障诊断 传感器优化配置 解析冗余关系 假定特征矩阵 整数规划 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]