基于改进人工蜂群算法的多值属性系统故障诊断策略  

Fault Diagnosis Strategy of Multi-Valued Attribute System Based on Improved Artificial Bee Colony Algorithm

作　　者：张志龙 陈岑 史贤俊 蒋勇 ZHANG Zhilong;CHEN Cen;SHI Xianjun;JIANG Yong(Naval Aviation University,Yantai Shandong 264001,China;Institute 706 of the Second Academy of CASIC,Beijing 100854,China)

机构地区：[1]海军航空大学,山东烟台264001 [2]中国航天科工集团第二研究院706所,北京100854

出　　处：《海军航空大学学报》2022年第3期261-268,共8页Journal of Naval Aviation University

基　　金：国家自然科学基金青年科学基金(61903374)。

摘　　要：针对实际工程应用中常见的多值属性系统故障诊断策略问题,在传统人工蜂群算法(Artificial Bee Colony Algorithm,简称ABC算法)的基础上,提出改进蜂群算法用于多值属性系统的测试序列寻优。首先,在蜂群算法中重新定义多值D矩阵和五元组的含义;其次,引入方向信息概率矩阵,设置状态转移规则和矩阵元素更新策略;最后,采用导弹舵系统实例说明算法的实现过程和有效性。最终测试序列寻优结果表明:与传统的多值Rollout算法相比,文中所提算法能得到较好的诊断策略结果,具有一定的应用价值。Aiming at the common fault diagnosis strategy problems of multi-valued attribute systems in practical engineering applications,based on the traditional artificial bee colony algorithm,an improved bee colony algorithm is proposed to optimize the test sequence of multi-valued attribute systems.Firstly,the meaning of the multi-valued D matrix and the five-tuple in the bee colony algorithm is redefined;secondly,the direction information probability matrix is introduced,the state transition rules and matrix element update strategies are set;finally,the missile rudder system example is used to illustrate the implementation process and the effectiveness of the algorithm.The final test sequence optimization results show that the algorithm proposed can obtain better diagnostic strategy results compared with the traditional multi-valued Rollout algorithm and has a certain application value.

关 键 词：多值属性系统 人工蜂群算法 故障诊断策略 导弹舵系统 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]