基于共识与Choquet积分的模糊Petri网可信度确定方法  

Method for determining credibility of fuzzy Petri net based on consensus and Choquet integration

作　　者：朱江洪[1,2] 李延来[1,2] 王睿[1,2] Zhu Jianghong;Li Yanlai;Wang Rui(School of Transportation ＆ Logistics;National United Engineering Laboratory of Integrated ＆ Intelligent Transportation,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学交通运输与物流学院,成都610031 [2]西南交通大学综合交通运输智能化国家地方联合工程实验室,成都610031

出　　处：《计算机应用研究》2018年第6期1700-1706,共7页Application Research of Computers

基　　金：国家自然科学基金资助项目(71371156)

摘　　要：针对模糊Petri网可信度确定过程中决策结果可接受程度较低和未考虑影响因子间关联的问题,提出基于群体共识测度与Choquet积分确定模糊Petri网可信度的方法。设计了基于决策者与群体评估信息偏差修正的自适应共识测度算法,获取可接受程度较高的群体评估信息。构建最大2-可加模糊测度Marichal熵模型求解影响因子间的交互作用系数,根据交互作用系数、默比乌斯变换和模糊测度之间的关系确定模糊测度,利用Choquet积分算子集结群体评估信息得到模糊Petri网可信度。将此方法用于确定燃气轮机故障诊断模糊Petri网输入库所可信度,并与未考虑共识测度和影响因子间关联的方法进行比较,计算结果验证了该方法的有效性和可行性。In the process of determining the credibility of fuzzy Petri net,the result of the decision making is lower and the correlation between the influence factors is unconsidered,this paper put forward a method of determining the fuzzy Petri net credibility based on group consensus measure and Choquet integral. It designed the adaptive consensus algorithm based on the decision maker and the group evaluation information error correction,and obtained the group evaluation information which could be accepted by the algorithm. It solved the interaction coefficient of the influence factors by the biggest 2-additve fuzzy measure Marichal entropy model. According to the relationship between the interaction coefficient,Mobius transform and fuzzy measure to determine the fuzzy measure,it obtained the fuzzy Petri net credibility by aggregating group evaluation information by using Choquet integral. It verified the validity and feasibility of this method by the case of determine the gas turbine fault diagnosis fuzzy Petri net input place credibility in comparison with a method of not consider the consensus measure and the correlation of influence factors.

关 键 词：模糊PETRI网 可信度 共识测度 模糊测度 CHOQUET积分 

分 类 号：TP307[自动化与计算机技术—计算机系统结构] C934[自动化与计算机技术—计算机科学与技术]