复杂机电系统关键部件辨识方法及应用

鉴于现有衡量系统中关键部件的重要性测度在构建时仅考虑拓扑重要性或功能重要性,本文利用Choquet积分,结合部件功能属性改进网络理论中拓扑重要性测度提出节点广义重要性测度的定义及普适性构建方法;针对单一重要性测度识别关键部件时可能存在的不确定性与随机性,提出聚合算子(aggregation operator,AO)融合多种广义重要性测度识别复杂机电系统中的关键部件.最后以高速列车转向架系统为例选取模糊积分为聚合算子AO识别系统中的关键部件,结果表明以模糊积分为聚合算子AO,同时利用k可加模糊测度确定权重的识别结果更为准确.

Identifying critical components of complex electromechanical system and its application

The topological importance or functional importance is only taken into consideration, when the existing measures are constructed. Based on functional properties of components, generalized importance measures, which improve topological measures in network theory, are proposed to combine with Choquet integral. In order to overcome the uncertainty and randomness, which exist if a single importance measure identifies critical components, aggregation operator (AO) is given through fusion multi-generalized importance measures and then used to identify critical components of complex electromechanical system. Finally, a case study which selects fuzzy integral to treat as AO is presented to identify critical components of bogie system. The results show that the method with Choquet integral regarding as AO and k fuzzy measure calculating weights, is effective.