摘要
针对传统方法在解决现代配电网故障分类时存在的求解速度慢、分类精度差的问题,提出一种结合量子遗传算法的支持向量机配电网故障分类方法。首先采集故障信号并进行S变换处理,将处理结果的均方根与均值作为特征量,以提高特征量辨识度;其次在量子遗传算法的种群更新策略中引入自适应动态旋转角,避免算法在早期陷入局部收敛,增加算法的调参精度。以IEEE33节点配电网模型为研究对象的实验结果表明,所提的特征量提取方法有效提高了支持向量机的故障分类精度,改进后的量子遗传算法通过为支持向量机寻到更优参数从而进一步提高了故障的判别精度和分类速度,验证了方法的有效性与准确性。
A support vector machine distribution network fault classification method combined with a quantum genetic algorithm is proposed as a way of overcoming the problems of slow solution speed and poor classification accuracy of traditional methods in modern distribution network fault classification. The fault signal is first collected and processed by S-transform, and the root mean square and mean value of the processing results are taken as the characteristic quantities to improve the identification degree of the characteristic quantities. An adaptive dynamic rotation angle is then introduced into the population update strategy of the quantum genetic algorithm in order to prevent the algorithm falling into local convergence in the early stages and increase the parameter adjustment accuracy of the algorithm. The experimental results for an IEEE 33 node distribution network model show that the proposed feature extraction method effectively increases the fault classification accuracy of support vector machines. The improved quantum genetic algorithm further improves the fault identification accuracy and classification speed by finding better parameters for the support vector machine, which verifies the effectiveness and accuracy of the method.
作者
罗程浩
赵启承
魏云冰
李思成
LUO ChengHao;ZHAO QiCheng;WEI YunBing;LI SiCheng(College of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620;Power Science Research Institute of Zhejiang Electric Power Co.,Ltd.,Hangzhou 310014,China)
出处
《北京化工大学学报（自然科学版）》
CAS
CSCD
北大核心
2022年第6期110-118,共9页
Journal of Beijing University of Chemical Technology(Natural Science Edition)
基金
国网浙江省电力有限公司科技项目(5211DS16001Y)
国家自然科学基金(51507157)。
关键词
故障分类
特征量提取
支持向量机
量子遗传算法
自适应动态旋转角
fault classification
feature extraction
support vector machine
quantum genetic algorithm
adaptive dynamic rotation angle
