Novel dynamic evidential Petri net for system reliability analysis

作     者：Wensheng Peng Jianguo Zhang Jinyang Zhang 

作者机构：School of Reliability and Systems Engineering Beihang University China Aero-Polytechnology Establishment Aviation Industry Corporation of China Science and Technology on Reliability and Environment Engineering Laboratory Beihang University 

出 版 物：《Journal of Systems Engineering and Electronics》 (系统工程与电子技术（英文版）)

年 卷 期：2017年第28卷第5期

页      面：1019-1027页

核心收录：

学科分类：0808[工学-电气工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 0802[工学-机械工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 081202[工学-计算机软件与理论] 

基　　金：supported by the National Basic Research Program of China(2013CB733002) 

主　　题：evidence theory Petri net knowledge representation improved ABC algorithm imprecise information reliability 

摘      要：This paper proposes a novel dynamic Petri net (PN) model based on Dempster-Shafer (D-S) evidence theory, and this improved evidential Petri net (EPN) model is used in knowledge inference and reliability analysis of complex mechanical systems. The EPN could take epistemic uncertainty such as interval information, subjective information into account by applying D-S evidence quantification theory. A dynamic representation model is also proposed based on the dynamic operation rules of the EPN model, and an improved artificial bee colony (ABC) algorithm is employed to proceed optimization calculation during the complex systems learning process. The improved ABC algorithm and D-S evidence theory overcome the disadvantage of extremely subjective in traditional knowledge inference efficiently and thus could improve the accuracy of the EPN learning model. Through a simple numerical case and a satellite driving system analysis, this paper proves the superiority of the EPN and the dynamic knowledge representation method in reliability analysis of complex systems.