基于并行隐马尔科夫模型的电能质量扰动事件分类

谢善益; 肖斐; 艾芊; 周刚

引用本文:	谢善益,肖斐,艾芊,等.基于并行隐马尔科夫模型的电能质量扰动事件分类[J].电力系统保护与控制,2019,47(2):80-86.
	XIE Shanyi,XIAO Fei,AI Qian,et al.Parallel hidden Markov model based classification of power quality disturbance events[J].Power System Protection and Control,2019,47(2):80-86

【打印本页】【下载PDF全文】【查看/发表评论】【EndNote】【RefMan】【BibTex】

←前一篇|后一篇→

过刊浏览高级检索

本文已被：浏览 6388次下载 2783次	码上扫一扫！
分享到：微信更多字体:加大+\|默认\|缩小-
基于并行隐马尔科夫模型的电能质量扰动事件分类
谢善益¹,肖斐²,艾芊²,周刚¹
(1.广东电网有限责任公司电力科学研究院，广东广州 510600;2.上海交通大学电气工程系，上海 200240)

摘要:

为满足电能质量扰动准确分类的需求，提出了一种基于极大重叠离散小波变换(Maximal Overlap Discrete Wavelet Transform, MODWT)和并行隐马尔科夫模型(Parallel Hidden Markov Model, PHMM)的电能质量扰动分类方法。首先利用MODWT提出一种实用的电能质量扰动检测算法，该算法无需设定检测阈值，可准确获取扰动时段的起止时刻。接着提取扰动时段的电压谐波成分并组成特征向量。然后用PHMM分类器对扰动信号进行分类识别。PHMM方法克服了人工神经网络方法收敛性较差、训练时间较长的缺陷，使分类器性能大大提升。通过应用于现场实测扰动数据表明，所提出的方法适用于多种类型的电能质量扰动检测，分类正确率高，训练速度快，具有良好的应用价值。

关键词: 电能质量极大重叠离散小波变换并行隐马尔科夫模型分类识别

DOI：10.7667/PSPC180062

分类号:

基金项目:广东电网公司科技项目资助(GDKJXM20162540);国家863计划课题项目资助(2015AA050404)

Parallel hidden Markov model based classification of power quality disturbance events

XIE Shanyi¹,XIAO Fei²,AI Qian²,ZHOU Gang¹

(1. Electric Power Research Institute of Guangdong Power Grid Co., Ltd., Guangzhou 510600, China;2. Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China)

Abstract:

In order to meet the requirements of accurately classifying power quality disturbances, a method for power quality disturbance classification is proposed based on Maximal Overlap Discrete Wavelet Transform (MODWT) and Parallel Hidden Markov Model (PHMM). Initially, a practical power quality disturbance detection algorithm is proposed by using MODWT. This algorithm can obtain the disturbance beginning and ending time accurately without setting detection threshold, from whose results the voltage harmonic components of power quality disturbance are extracted and used to form feature vector. Then, PHMM, as a classifier, is used to identify power quality disturbances. PHMM method solves the problem of poor convergence and longer training time for Artificial Neural Network (ANN) method, and thus the performance of the classifier is greatly improved. The test results based on power grid field data show that the proposed method is suitable for detecting various types of power quality disturbances, and it is characterized by high recognition correctness and less training time, and it will find extensive application. This work is supported by Science and Technology Project of Guangdong Power Grid Company (No. GDKJXM20162540) and National High-tech R & D Program of China (863 Program) (No. 2015AA050404).

Key words: power quality maximum overlapping discrete wavelet transform parallel hidden Markov model classification and identification

X关闭