基于可靠度和一致强度的冲突证据组合算法

A conflicting evidence combination algorithm based on reliability and coherence intensity

考虑到利用无线传感器网络对天然气管道进行泄漏监控时,不同传感器节点的诊断结果之间会发生冲突,致使直接利用Dempster-Shafer(D-S)组合规则或其修正组合规则进行冲突证据组合的方法难以让汇聚节点做出正确决策,为此提出了一种基于可靠度和一致强度的冲突证据组合算法——CECARCI。该算法首先利用各监测节点的可靠度对证据集进行预处理,减小不可靠证据对组合结果的影响,然后通过对所有证据进行整体分析,引入了证据一致强度和基元支持度,调整证据组合时的次序,并以合理的权重来分配证据间的冲突。算例实验结果表明,相比于D-S组合法则、Yager组合法则和Murphy组合法则,CECARCI算法可获得收敛性更好的组合结果。

Considering that when wireless sensor networks are applied to security monitoring for gas pipelines, the diagnosis results from different sensor nodes may seriously conflict, so it is difficult for the sink node to make correct decisions using the Dempster-Shafer （D-S） Combination rule or its modified combination rules, this paper proposes a new conflicting evidence combination algorithm based on reliability and coherence intensity （CECARCI）. Firstly, the algorithm preproeesses the evidence set according to nodes＇ reliability in order to decrease the effect of unreliable evidences on the fusion result, and then analyzes all the evidences, and obtains the sequence of combination according to the coherence intensity of evidence. Additionally, the coherence intensity and the support degree of base element are used to manage the conflict. The numerical example shows that the CECARCI can obtain more reasonable results with good convergence compared with D- S, Yager or Murphy combination rules.