基于ABC-DNN的小电流接地故障选线方法
|
摘要
为了提高小电流接地系统故障选线的速度和精度,使系统快速准确地进行故障选线,提出了基于人工蜂群优化深度神经网络(ABC-DNN)的故障选线方法。根据杭州某配电系统的实际数据,利用Matlab/Simulink搭建了小电流接地系统模型,获取故障线路中的零序电流,并从中提取暂态能量分量、稳态基波分量以及五次谐波分量作为样本数据,输入经人工蜂群优化的深度神经网络模型,经过训练输出选线结果。通过人工蜂群算法优化网络的权重,在一定程度上缩短了训练时间,提高了判断准确性。根据实际数据得到的选线结果显示,该方法降低了训练时间、提高了判断精度,并且对系统拓扑结构具有鲁棒性,能够达到实际应用的要求。
In order to improve the speed and accuracy of fault line selection in small current grounding power system, and make the system fault line selection quickly and accurately,a fault line selection method based on ABC-DNN(deep neural network optimized by artificial bee colony algorithm) is proposed. The Matlab/Simulink is used to build a small current grounding system model to obtain the zero-sequence current in the faulty line, and the transient energy component, steady-state fundamental component and steady-state fifth harmonic component are extracted as sample data. These data are put into a deep neural network model optimized by artificial bee colony, and are trained to output the line selection results. Optimizing the weight of the network by the artificial bee colony algorithm can shorten the training time to a certain degree and improves the accuracy. The simulation results show that the method can accelerate the learning training speed and improve the judgment accuracy, and the system topology is robust and can meet the requirements of practical fault line selection.
In order to improve the speed and accuracy of fault line selection in small current grounding power system, and make the system fault line selection quickly and accurately,a fault line selection method based on ABC-DNN(deep neural network optimized by artificial bee colony algorithm) is proposed. The Matlab/Simulink is used to build a small current grounding system model to obtain the zero-sequence current in the faulty line, and the transient energy component, steady-state fundamental component and steady-state fifth harmonic component are extracted as sample data. These data are put into a deep neural network model optimized by artificial bee colony, and are trained to output the line selection results. Optimizing the weight of the network by the artificial bee colony algorithm can shorten the training time to a certain degree and improves the accuracy. The simulation results show that the method can accelerate the learning training speed and improve the judgment accuracy, and the system topology is robust and can meet the requirements of practical fault line selection.
引文
[1]潘本仁,宋华茂,张秋凤,等.小电流接地故障无功功率分析及选线新方法[J].电力系统保护与控制,2017, 45(14):51-56.PAN Benren, SONG Huamao. ZHANG Qiufeng, et al.Reactive power analysis and novel faulty selection method in resonant grounding system[J]. Power System Protection and Control, 2017, 45(14):51-56.
[2]丁爱佳,鲍炫羽,季玲玲,等.小电流接地系统单相接地故障选线方法研究[J].浙江电力,2017, 36(5):8-11.DING Aijia, BAO Xuanyu, JI Lingling, et al. Research on single phase-to-ground fault line selection methods in small current grounding system[J]. Zhejiang Electric Power, 2017, 36(5):8-11
[3]王磊,曹现峰,骆玮.基于改进PSO优化模糊神经网络的配电网故障选线研究[J].电气技术,2016, 17(3):30-35.WANG Lei, CAO Xianfeng, LUO wei. Research on fault line detection for distribution network based on improved pso to optimize fuzzy neural network[J]. Electrical Engineering, 2016,17(3):30-35.
[4]高杰,程启明,程尹曼,等.基于量子遗传双稳态系统的配电网故障选线方法[J].电力自动化设备,2018,38(5):164-170.GAO Jie, Cheng Qiming, CHENG Yinman, et al. Faulty line selection method based on quantum genetic bistable system for distribution network[J]. Electric Power Automation Equipment, 2018, 38(5):164-170.
[5]张姝,何正友,林圣,等.暂态零序电荷-电压特征与支持向量机结合的谐振接地系统故障选线研究[J].电力系统保护与控制,2013,12(8),24-27.ZHANG Zhu, HE Zhengyou, LIN Sheng, et al. Fault line selection of resonant grounding system based on the characteristics of charge-voltage in the transient zero sequence and support vector machine[J]. Power System Protection and Control, 2013, 12(8):51-56.
[7]徐铭铭,高淑萍,常仲学,等.基于模型识别的消弧线圈接地系统单相接地选线方法[J].电力系统保护与控制,2018,46(2):73-77.XU Mingming, GAO Shuping, CHANG Zhongxue, et al.Model recognition based single-phase earthed faulty line selection method of Petersen coil grounded system[J].Power System Protection and Control, 2018, 46(2):73-77.
[8]薛太林,侯隽朗,张建新.基于GA优化BP神经网络小电流接地系统故障选线方法[J].电气自动化,2018,40(2):66-69.XUE Tailin, HOU Junlang, ZHANG Jianxin. Fault line detection in low current grounding system based on GA optimized BP neural network[J]. Electrical Automation2018,40(2):66-69.
[9]袁琴琴,吕林涛.基于改进蚁群算法与遗传算法组合的网络人侵检测[J].重庆邮电大学学报(自然科学版),2017, 29(1):84-89.YUAN Qinqin, LU Lintao. Network intrusion detection method based on combination of improved ant colony optimization and genetic algorithm[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2017, 29(1):84-89.
[10]孟令通,朱洪渊,蒋祖华,等.基于遗传算法的平板车调度优化方法[J].哈尔滨工程大学学报,2018, 39(3):554-560.MENG Lingtong, ZHU Hongyuan, JIANG Zuhua, el al.Flat-car scheduling optimization method based on genetic algorithm[J]. Journal of Harbin Engineering University,2018, 39(3):554-560.
[11]高卫峰,刘三阳,黄玲玲.受启发的人工蜂群算法在全局优化问题中的应用[J1.电子学报,2012, 40(12):2396-2403.GAO Weifeng, LIU Sanyang, HUANG Lingling. Inspired artificial bee colony algorithm for global optimization problems[J]. Acta Electronica Sinica, 2012, 40(12):2396-2403.
[12]吴乐鹏,黄纯,林达斌,等.基于暂态小波能量的小电流接地故障选线新方法[J].电力自动化设备,2013, 33(5):70-75.WU Lepeng, HUANG Chun, LIN Dabin, et al. Faulty line selection based on transient wavelet energy for nonsolid-earthed network[J]. Electric Power Automation Equipment,2013,33(5):70-75.
[13]牛勃,丁培,马飞越,等.粒子群优化算法在电气设备特高频局部放电源定位中的应用[J].智慧电力,2018,46(9):101-108.NIU Bo, DING Pei, MA Feiyue, et al. Application of partice swarm optimization algorithm in electrical equipment UHF PD source positioning[J]. Smart Power,2018, 46(9):101-108.
[14]张弘喆,赵康,李乐蒙,等.基于多层人工神经网络的电力系统恢复过电压预测[J].智慧电力,2018, 46(8):67-73.ZHANG Hongzhe, ZHAO Kang, LI Lemeng, el al.Overvoltage forecast during power system restoration based on multilayer artificial neural network[J]. Smart Power,2018, 46(8):67-73..
[15]王庆斌.石亮缘,黄辉,等.基于改进粒子群算法的输电线路路径规划研究[J].广东电力,2018, 31(9):142-148.WANG Qingbin, SHI Liangyuan, HUANG Hui, et al.Research on power transmission line path planning based on improved particle swarm optimization algorithm[J].Guangdong Electric Power, 2018, 31(9):142-148.
[16]康小平,王溢熹,陈德炜,等.配电网大面积停电时快速恢复路径选择算法研究[J].电力信息与通信技术,2018, 16(4):64-69.KANG Xiaoping, WANG Yixi, CHENG Dewei, el al.Research on fast recovery path selection algorithm of large blackout in distribution network[J]. Electric PowerInformation and Communication Technology, 2018, 16(4):64-69.
[17]郭昱慧.基于模型的BP网络辨识算法代码快速生成[J].电子设计工程,2018, 26(12):187-191.GUO Yuhui. Model based BP network identification algorithm fast code generation[J]. Electronic Design Engineering, 2018, 26(12):187-191.
[2]丁爱佳,鲍炫羽,季玲玲,等.小电流接地系统单相接地故障选线方法研究[J].浙江电力,2017, 36(5):8-11.DING Aijia, BAO Xuanyu, JI Lingling, et al. Research on single phase-to-ground fault line selection methods in small current grounding system[J]. Zhejiang Electric Power, 2017, 36(5):8-11
[3]王磊,曹现峰,骆玮.基于改进PSO优化模糊神经网络的配电网故障选线研究[J].电气技术,2016, 17(3):30-35.WANG Lei, CAO Xianfeng, LUO wei. Research on fault line detection for distribution network based on improved pso to optimize fuzzy neural network[J]. Electrical Engineering, 2016,17(3):30-35.
[4]高杰,程启明,程尹曼,等.基于量子遗传双稳态系统的配电网故障选线方法[J].电力自动化设备,2018,38(5):164-170.GAO Jie, Cheng Qiming, CHENG Yinman, et al. Faulty line selection method based on quantum genetic bistable system for distribution network[J]. Electric Power Automation Equipment, 2018, 38(5):164-170.
[5]张姝,何正友,林圣,等.暂态零序电荷-电压特征与支持向量机结合的谐振接地系统故障选线研究[J].电力系统保护与控制,2013,12(8),24-27.ZHANG Zhu, HE Zhengyou, LIN Sheng, et al. Fault line selection of resonant grounding system based on the characteristics of charge-voltage in the transient zero sequence and support vector machine[J]. Power System Protection and Control, 2013, 12(8):51-56.
[7]徐铭铭,高淑萍,常仲学,等.基于模型识别的消弧线圈接地系统单相接地选线方法[J].电力系统保护与控制,2018,46(2):73-77.XU Mingming, GAO Shuping, CHANG Zhongxue, et al.Model recognition based single-phase earthed faulty line selection method of Petersen coil grounded system[J].Power System Protection and Control, 2018, 46(2):73-77.
[8]薛太林,侯隽朗,张建新.基于GA优化BP神经网络小电流接地系统故障选线方法[J].电气自动化,2018,40(2):66-69.XUE Tailin, HOU Junlang, ZHANG Jianxin. Fault line detection in low current grounding system based on GA optimized BP neural network[J]. Electrical Automation2018,40(2):66-69.
[9]袁琴琴,吕林涛.基于改进蚁群算法与遗传算法组合的网络人侵检测[J].重庆邮电大学学报(自然科学版),2017, 29(1):84-89.YUAN Qinqin, LU Lintao. Network intrusion detection method based on combination of improved ant colony optimization and genetic algorithm[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2017, 29(1):84-89.
[10]孟令通,朱洪渊,蒋祖华,等.基于遗传算法的平板车调度优化方法[J].哈尔滨工程大学学报,2018, 39(3):554-560.MENG Lingtong, ZHU Hongyuan, JIANG Zuhua, el al.Flat-car scheduling optimization method based on genetic algorithm[J]. Journal of Harbin Engineering University,2018, 39(3):554-560.
[11]高卫峰,刘三阳,黄玲玲.受启发的人工蜂群算法在全局优化问题中的应用[J1.电子学报,2012, 40(12):2396-2403.GAO Weifeng, LIU Sanyang, HUANG Lingling. Inspired artificial bee colony algorithm for global optimization problems[J]. Acta Electronica Sinica, 2012, 40(12):2396-2403.
[12]吴乐鹏,黄纯,林达斌,等.基于暂态小波能量的小电流接地故障选线新方法[J].电力自动化设备,2013, 33(5):70-75.WU Lepeng, HUANG Chun, LIN Dabin, et al. Faulty line selection based on transient wavelet energy for nonsolid-earthed network[J]. Electric Power Automation Equipment,2013,33(5):70-75.
[13]牛勃,丁培,马飞越,等.粒子群优化算法在电气设备特高频局部放电源定位中的应用[J].智慧电力,2018,46(9):101-108.NIU Bo, DING Pei, MA Feiyue, et al. Application of partice swarm optimization algorithm in electrical equipment UHF PD source positioning[J]. Smart Power,2018, 46(9):101-108.
[14]张弘喆,赵康,李乐蒙,等.基于多层人工神经网络的电力系统恢复过电压预测[J].智慧电力,2018, 46(8):67-73.ZHANG Hongzhe, ZHAO Kang, LI Lemeng, el al.Overvoltage forecast during power system restoration based on multilayer artificial neural network[J]. Smart Power,2018, 46(8):67-73..
[15]王庆斌.石亮缘,黄辉,等.基于改进粒子群算法的输电线路路径规划研究[J].广东电力,2018, 31(9):142-148.WANG Qingbin, SHI Liangyuan, HUANG Hui, et al.Research on power transmission line path planning based on improved particle swarm optimization algorithm[J].Guangdong Electric Power, 2018, 31(9):142-148.
[16]康小平,王溢熹,陈德炜,等.配电网大面积停电时快速恢复路径选择算法研究[J].电力信息与通信技术,2018, 16(4):64-69.KANG Xiaoping, WANG Yixi, CHENG Dewei, el al.Research on fast recovery path selection algorithm of large blackout in distribution network[J]. Electric PowerInformation and Communication Technology, 2018, 16(4):64-69.
[17]郭昱慧.基于模型的BP网络辨识算法代码快速生成[J].电子设计工程,2018, 26(12):187-191.GUO Yuhui. Model based BP network identification algorithm fast code generation[J]. Electronic Design Engineering, 2018, 26(12):187-191.