基于简化因果图的工业过程故障根本变量诊断  

Root variable diagnosis of industrial process fault based on simplifying causality diagram

作　　者：郭小萍 洪升园 李元 Guo Xiaoping;Hong Shengyuan;Li Yuan(College of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China)

机构地区：[1]沈阳化工大学信息工程学院,沈阳110142

出　　处：《计算机应用研究》2022年第12期3720-3723,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61673279);辽宁省教育厅资助项目(LJ2020021)。

摘　　要：在对工业过程故障进行根本原因诊断时,由于过程的自身特性和反馈控制等因素的干扰,使得变量因果图过于复杂从而使故障传播路径难以解释且不能找到导致故障的根本变量。提出一种简化因果图的方法,通过两步走对收敛交叉映射法构建的因果图实现简化,保留主要的故障传播路径。首先采用模糊综合评判法判别因果图中不确定性的关系;然后通过求解最大生成树,得到赋权无向图,并根据变量间因果关系选取根节点,分析赋权无向图获得新路径,从而将其改进成赋权有向图。在田纳西—伊斯曼过程进行验证实验,并与传统收敛交叉映射法进行比较,结果验证了所提出方法的有效性。When root cause diagnosis for industrial process fault,due to the characteristics of the process itself and feedback control,the causality diagram of variables is too complicated,so that the fault propagation path is difficult to explain and unable to find the most fundamental variable leading to the fault.This paper proposed a method of simplifying the causality diagram.It could simplify the causality diagram constructed by the convergent cross mapping method through“two steps”and preserved the main fault propagation paths.Firstly,the fuzzy comprehensive evaluation determined the relationship of uncertainty in the causality diagram.Then it got the weighted undirected graph by solving the maximum spanning tree.Selecting the root node based on the causal relationship between variables,it obtained new path and the weighted undirected graph to improve it into the weighted directed graph.This paper did the validation experiments in the TE process.Compared with the traditional convergent cross mapping method,the results show that the proposed method is effective.

关 键 词：根本原因诊断 简化因果图 收敛交叉映射 模糊综合评判 最大生成树 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]