双因果键合图在传感器布局方法中的应用研究

传感器布置方案在满足监测性能的同时,也应该考虑系统故障诊断与健康状态监测的需求。目前常用的传感器配置方法主要依赖系统解析冗余关系(ARRs),而ARRs通常是非常复杂难以推导。因此,提出了一种基于双因果键合图理论的路径分析法,通过因果路径关系构建系统的故障特征矩阵(FSM),在确保系统故障可诊断和可隔离性的基础上,得到最优传感器配置方案。以两容水箱系统为例,建立系统双因果键合图模型,在隐式ARRs的基础上得到系统FSM,推导得到系统最优传感器配置方案,并进行了推导试验验证,确保了所提方法的可靠性。

Application of Bicausal Bond Graph in Sensor Placement Method

Sensor placement has a great influence on the fault diagnosis characteristics of a system. Aiming at the difficulty of deriving analytic redundancy relation(ARR) for specific mathematical expression in traditional sensor placement method based on system model, a causal path analysis method based on bicausal bond graph model is proposed. The fault characteristic matrix(FSM) of the system was constructed by analyzing the causal path relationship of the bicausal bond graph model. On the basis of guaranteeing the isolation of system faults, the optimal sensor configuration scheme was obtained. Finally, the bond graph and bicausal bond graph model of a two tank system were established, and the optimal sensor configuration scheme was obtained through the proposed method. The proposed optimal sensor configuration scheme was deduced and verified to ensure the fault isolation of the main components of the system.