基于蜂群优化核极限学习机的电能扰动识别方法  

Recognition Method of Electric Energy Disturbance Based on Nuclear Extreme Learning Machine of Bee Colony Optimization

作　　者：何昌龙 曲丽萍 张杰 高泰路 HE Changlong;QU Liping;ZHANG Jie;GAO Tailu(College of Electrical and Information Engineering,Beihua University,Jilin 132021,China)

机构地区：[1]北华大学电气与信息工程学院,吉林吉林132021

出　　处：《北华大学学报（自然科学版）》2022年第2期265-273,共9页Journal of Beihua University（Natural Science）

基　　金：国家重点新产品计划项目(2010GRB10003);吉林省科技发展计划项目(20190102015JH);吉林省教育厅科学技术研究项目(JJKH20200043KJ).

摘　　要：针对电能质量扰动信号分类识别的需要,以及BP(Back Propagation)网络分类器模型计算量大、训练时间长、计算速度无法满足电力系统在线分析要求等问题,提出一种蜂群优化核极限学习机分类器模型,采用小波变换提取电能质量扰动信号的特征向量作为样本,使用最优参数下的核极限学习机分类识别电能质量扰动信号.仿真结果表明:与未优化极限学习机相比,蜂群优化核极限学习机分类器分类识别的正确率提高了20%,误判率降低3%,蜂群优化核极限学习机分类模型对电能质量扰动识别具有一定的效果.In view of the need for classification and recognition of power quality disturbance signals,as well as the problems of Back Propagation network classifier model with large calculation amount,long training time,and calculation speed that can not meet the requirements of power system online analysis,this paper proposes a bee colony optimization kernel extreme learning.The machine classifier model uses wavelet transform to extract the characteristic vector of the power quality disturbance signal as a sample,and uses the kernel extreme learning machine under the optimal parameters to classify and recognize the power quality disturbance signal.Simulation results show:after the network optimized by the bee colony algorithm,the correct rate of classification and recognition has increased by nearly 20%,and the misjudgment rate has dropped to 3%.The use of bee colony algorithm optimizes the recognition of power quality disturbances by the nuclear extreme learning machine.

关 键 词：电能扰动识别 小波变换 电能扰动的能量表征 核极限学习机 蜂群优化 

分 类 号：TM715[电气工程—电力系统及自动化]