基于改进证据理论的冲突证据融合方法  

Conflict evidence fusion method based on improved evidence theory

作　　者：张斌 侯紫霞 张建锋[1] ZHANG Bin;HOU Zi-xia;ZHANG Jian-feng(College of Information Engineering,Northwest A&F University,Yangling 712100,China)

机构地区：[1]西北农林科技大学信息工程学院,陕西杨凌712100

出　　处：《计算机工程与设计》2024年第6期1683-1690,共8页Computer Engineering and Design

基　　金：陕西省重点研发计划基金项目(2023-YBNY-212)。

摘　　要：为解决证据理论中Dempster组合规则融合高度冲突证据效果不佳的问题,提出一种同时考虑证据重要性与可靠性的综合权重生成方法和基于加权复数基本概率分配函数的证据融合方法。以提高证据融合结果的收敛性能和诊断精度为目标,建立相应的优化证据融合模型。实验结果表明,所提方法能够有效识别并处理证据冲突,克服证据加权融合过程中剩余信度被完全抛弃的问题,获得更好的融合结果和更快的收敛速度,提高诊断精度。To address the problem of poor performance caused by highly conflicting evidence in the Dempster combination rule fusion of evidence theory,a comprehensive weight generation method that considered both the importance and reliability of evidence,and an evidence fusion method based on weighted complex basic probability assignment function were proposed.A corresponding optimized evidence fusion model was established with the goal of improving the convergence performance and diagnostic accuracy of evidence fusion.Experimental results show that the proposed method can effectively identify and handle evidence conflicts,overcome the problem of completely discarding residual credibility in the evidence weighting fusion process,and achieve better fusion results and higher convergence speed,effectively improving diagnostic accuracy.

关 键 词：信息融合 多准则决策 证据理论 冲突证据融合 证据融合规则 证据综合权重 置信分布 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]