基于CWOA-BP的电缆接头阻水性能评估  

Water Resistance Performance Evaluation of Cable Joints Based on CWOA-BP

作　　者：吕峰 何光华 王昱力 卞栋 庄裕 陈馨凝 LV Feng;HE Guang-hua;WANG Yu-li;BIAN Dong;ZHUANG Yu;CHEN Xin-ning(Wuxi Power Supply Branch,State Grid Jiangsu Electric Power Co., Ltd., Wuxi 214061,China;Wuhan Branchof China Electric Power Research Institute Co.,Ltd., Wuhan 430074, China;Hubei Institute of Power Grid Intelligent Control and Equipment Engineering Technology Research Center, Hubei University of Technology , Wuhan 430068 ,China)

机构地区：[1]国网江苏省电力有限公司无锡供电分公司,江苏无锡214061 [2]中国电力科学研究院有限公司武汉分院,湖北武汉430074 [3]湖北工业大学湖北省电网智能控制与装备工程技术研究中心,湖北武汉430068

出　　处：《水电能源科学》2019年第8期177-179,195,共4页Water Resources and Power

基　　金：国网江苏省电力有限公司科技项目（GYW11201702523）

摘　　要：为评估电缆接头阻水性,提出了一种基于CWOA-BP神经网络的电缆接头阻水性评估模型。对电缆冷缩接头进行阻水试验,选取电缆接头传感器上各项数据作为评估指标进行实例分析,并与BP、GA-BP、PSO-BP三种算法进行对比。仿真结果表明,新模型算法具有收敛速度快、预测精度高的优点,是一种有效可靠的电缆接头阻水性能评估模型。To evaluate the water resistance of cable joints, this paper proposes a water resistance evaluation model for cable joints based on CWOA-BP neural network. The water-resistance experiment of cable cold-shrink joints is conducted. The data on the cable joint sensor are selected as an evaluation index for example analysis, and compared with the BP, GA-BP and PSO-BP. The simulation results show that the algorithm adopted by the new model has the advantages of fast convergence speed and high prediction accuracy. Thus, it is an effective and reliable evaluation model for cable joint water resistance performance.

关 键 词：电缆接头 人工神经网络 混沌鲸群算法 评估模型 阻水性能 

分 类 号：TM855[电气工程—高电压与绝缘技术]