基于精英策略和ABBO算法的输电线路故障诊断模型  

Fault Diagnosis Model of Transmission Line Based on Elite Strategy and ABBO Algorithm

作　　者：冯艳敏 白梅娟 侯帅[1] 裴东锋 孙彭帅 秦亚洲[3] FENG Yan-min;BAI Mei-juan;Hou Shuai;PEI Dong-feng;SUN Peng-shuai;QIN Ya-zhou(Department of Information and Electrical Engineering,Hebei University of Engineering,Handan 056000,China;STATE Grid Handan Electric Power Supply Company,Handan 056000,China;China North Vehicle Research Institute,Beijing 100072,China)

机构地区：[1]河北工程大学信息与电气工程学院,河北邯郸056038 [2]国网河北省电力有限公司邯郸供电分公司,河北邯郸056002 [3]中国北方车辆研究所,北京100072

出　　处：《电脑与信息技术》2023年第5期48-52,共5页Computer and Information Technology

基　　金：国网河北省电力有限公司科技项目(项目编号:kj2021-042);国家自然科学基金项目(项目编号:61802107)。

摘　　要：针对输电线路故障分类问题,提出了一种基于精英策略的生物地理学与算术优化混合算法(ABBO)的输电线路故障诊断模型。首先,为了避免BBO算法的早熟收敛,提出将算术优化算法(AOA)的乘除加减算子代替生物地理学优化算法(BBO)的变异算子以优化特征属性权重;其次,提出将低于精度阈值的非精英个体进行二次变异的精英策略,以保留优质个体的方式提高算法精度;最后,提出一种基于重力的固定半径的近邻检索方式以改善案例推理中存在的不平衡问题。实验证明,与传统的基于BBO、AOA算法的故障诊断模型相比,该模型能够更加准确的识别故障类型。Aiming at the problem of transmission line fault classification,a hybrid algorithm of biogeography and arithmetic optimization(ABBO)based on elite strategy is proposed.Firstly,in order to avoid premature convergence of BBO algorithm,the mutation operator of Biogeography-Based Optimization(BBO)algorithm is replaced by multiplication,division,addition and subtraction operator of Arithmetic Optimization Algorithm(AOA)to optimize the weights of feature attributes.Secondly,an elite strategy of secondary mutation of non-elite individuals below the accuracy threshold is proposed to improve the accuracy of the algorithm by retaining high-quality individuals.Finally,a fixed radius neighbor retrieval method based on gravity is proposed to improve the imbalance problem of nearest neighbor classification in case-based reasoning.Experiments show that compared with the traditional fault diagnosis model based on BBO and AOA algorithms,this model can identify fault types more accurately.

关 键 词：精英策略 BBO 案例推理 故障诊断 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]