基于约束熵的复杂系统重要节点崩溃控制研究  

Complex System Important Node Collapse Control Based on Constraint Entropy

作　　者：汪送[1,2] 王瑛[1] 冯建伟[2] 杜纯[1] 陈盖凯[1] 

机构地区：[1]空军工程大学工程学院,陕西西安710038 [2]武警工程大学装备工程学院,陕西西安710086

出　　处：《中国安全科学学报》2012年第5期10-16,共7页China Safety Science Journal

基　　金：国家自然科学基金资助(71171199)

摘　　要：为有效控制复杂系统崩溃,提出一种基于约束熵的复杂系统重要节点崩溃控制方法。通过概述现有事故致因理论及模型,建立分析复杂系统安全事故致因的"认知-约束"模型。针对复杂系统节点众多,关联复杂的本质,指出为预防安全事故,必须在强化安全认知能力的基础上对重要节点进行约束控制。在提取节点重要度评估参数后,设计一种改进的复杂系统节点重要度评估算法。最后,引入约束熵概念,构建抑制复杂系统重要节点崩溃的约束熵度量模型。研究结果表明,存在可量化的用于控制复杂系统崩溃的手段;预防安全事故的关键是将适量的约束熵"引入"重要节点。In order to effectively control the collapse of complex system can be controlled effectively, a constraint entropy-based method for controlling complex system important node collapse was proposed. After summarizing the existing theories and models about accident causation , a cognition-constraint model was built for analyzing the safety accident causation of complex system . For numerous nodes and associa- tions exist in complex system, it was pointed out that in order to prevent safety accident , importance nodes in the complex system must be constrained and controlled on the basis of increasing safety cognition abili- ty. After the evaluation parameters of node importance were extracted, designed an improved complex system node important evaluation algorithm. Finally , the concept of constraint entropy was introduced. A constraint entropy measurement model was built. Research results show that there is a quantifiable means for controlling complex system collapse . The key to preventing safety accident is ＂introducing＂ a right amount of constraint entropy into the important node.

关 键 词：复杂系统 约束熵 节点重要度 “认知-约束”模型 崩溃控制 

分 类 号：X928.02[环境科学与工程—安全科学] N941.4[自然科学总论—系统科学]