基于PSO寻优与DBN神经网络的电晕损耗预测

doi:10.11930/j.issn.1004-9649.202112041

摘要/Abstract

摘要： 特高压交流输电线路的电晕损耗与降雨量、比湿、温度、相对湿度、压强等天气条件有相关性，可通过部分天气条件对特高压交流输电线路电晕损耗进行预测，提出了一种特高压交流输电线路的电晕损耗预测方法。根据粒子群优化算法（particle swarm optimization，PSO）寻优机制与深度信念网络（deep belief network，DBN）预测原理，详细说明了该预测方法的智能算法机制，并提出了一套完整的基于PSO—DBN智能算法的预测方法。首先，通过斯皮尔曼相关系数的大小确定与电晕损耗有较强相关性的天气条件，并作为特征值；然后以所选特征值为指标体系构建DBN神经网络进行电晕损耗预测，再采用PSO寻优算法对DBN神经网络进行内部参数调整，提升DBN神经网络的预测准确性；最后利用所提算法对实际运行的闽浙特高压输电线路的电晕损耗进行算法预测，与该线路的运行统计电晕损耗值进行对比分析，验证了所提预测方法的可行性。该方法为特高压输电线路电晕损耗研究和工程设计提供参考。

关键词: 特高压, 电晕损耗, 斯皮尔曼相关系数, 粒子群优化算法, DBN神经网络

Abstract: Based on the correlation between corona loss of UHV AC transmission lines and weather conditions such as rainfall, specific humidity, temperature, relative humidity, pressure, etc., a corona loss prediction method for UHV AC transmission lines is proposed by predicting corona loss of UHV AC transmission lines under some weather conditions. Based on the optimization mechanism of particle swarm optimization (PSO) and the prediction principle of deep belief network (DBN), the intelligent algorithm mechanism of this prediction method is explained in detail, and a complete set of prediction methods based on PSO—DBN intelligent algorithm is proposed. Firstly, the weather conditions that have strong correlation with corona loss are determined by the size of the Spearman correlation coefficient and used as the characteristic values. Then the DBN neural network is built with the selected eigenvalues as the index system to predict the corona loss. Then the internal parameters of the DBN neural network are adjusted by PSO optimization algorithm to improve the prediction accuracy of the DBN neural network. Finally, the proposed algorithm is used to predict the corona loss of the actual Fujian—Zhejiang UHV transmission line, and the statistical corona loss values of the line are compared and analyzed to verify the feasibility of the proposed prediction method. This method provides reference for corona loss research and engineering design of extra high voltage transmission lines.

Key words: UHV, corona loss, Spearman correlation coefficient, particle swarm optimization, deep belief network

黄书民, 蒋林高, 李志川, 杨光绪, 宋福根. 基于PSO寻优与DBN神经网络的电晕损耗预测[J]. 中国电力, 2022, 55(6): 95-102,214.

HUANG Shumin, JIANG Lingao, LI Zhichuan, YANG Guangxu, SONG Fugen. Corona Loss Prediction of UHV AC Transmission Line Based on DBN Neural Network Optimized by PSO[J]. Electric Power, 2022, 55(6): 95-102,214.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202112041

https://www.electricpower.com.cn/CN/Y2022/V55/I6/95

参考文献

[1] 杨建华, 肖达强, 张伟, 等. 基于改进PSO优化的RBF神经网络的特高压线损预测[J]. 广东电力, 2020, 33(9): 85–91
YANG Jianhua, XIAO Daqiang, ZHANG Wei, et al. Prediction of UHV line loss based on RBFNN optimized by improved PSO[J]. Guangdong Electric Power, 2020, 33(9): 85–91
[2] YIN F H, FARZANEH M, JIANG X L. Corona investigation of an energized conductor under various weather conditions[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2017, 24(1): 462–470.
[3] 刘健犇, 张波, 徐鹏飞. 雨天交流输电线路电晕放电判据[J]. 智慧电力, 2020, 48(4): 15–20, 35
LIU Jianben, ZHANG Bo, XU Pengfei. Corona discharge criterion for AC transmission lines in the rain[J]. Smart Power, 2020, 48(4): 15–20, 35
[4] 高晨光. 基于电晕笼和试验线段的高压直流输电电晕损耗等效性研究[D]. 北京: 华北电力大学, 2019: 1–2.
GAO Chenguang. Research of equivalence of HVDC corona loss of the conductors based on corona cage and test line[D]. Beijing: North China Electric Power University, 2019: 1–2.
[5] LIU L P, ZHANG J, WANG Q, et al. Theoretical calculation and evaluation of the line losses on UHV AC demonstration project[C]//2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems. Shenyang, China. IEEE, 2015: 1299–1303.
[6] 赵光锋, 李欣唐, 聂钢, 等. 基于电晕损耗计算的特高压交流同塔双回输电线路损耗特性[J]. 科学技术与工程, 2018, 18(30): 177–182
ZHAO Guangfeng, LI Xintang, NIE Gang, et al. Loss characteristics of ultra high voltage alternating current dual circuit transmission line based on corona loss calculation[J]. Science Technology and Engineering, 2018, 18(30): 177–182
[7] 刘云鹏, 黄世龙, 陈思佳, 等. 同塔四回750 kV六层横担输电线路电晕损失研究[J]. 高电压技术, 2019, 45(4): 1118–1123
LIU Yunpeng, HUANG Shilong, CHEN Sijia, et al. Corona loss of 750 kV four-circuit lines with six-layer cross-arms on the same tower[J]. High Voltage Engineering, 2019, 45(4): 1118–1123
[8] 王江储. 基于数据知识融合的输电线路线损计算与线损异常分析研究[D]. 广州: 华南理工大学, 2020: 16–21.
WANG Jiangchu. Research on line loss calculation and line loss anomaly analysis of transmission line based on data knowledge fusion[D]. Guangzhou: South China University of Technology, 2020: 16–21.
[9] 黄瑛. 计及电晕和谐波的500 kV电力网电能损耗计算[D]. 南京: 河海大学, 2005: 20–29.
HUANG Ying. A calculation of energy losses with consideration of corona losses and harmonic energy losses in 500 kV grids[D]. Nanjing: Hohai University, 2005: 20–29.
[10] 梁宇镔, 张耀宇, 谭家祺, 等. 基于随机矩阵理论的配电网状态分析与故障定位[J]. 电网与清洁能源, 2021, 37(3): 65–70
LIANG Yubin, ZHANG Yaoyu, TAN Jiaqi, et al. State analysis and fault location of distribution networks based on random matrix theory[J]. Power System and Clean Energy, 2021, 37(3): 65–70
[11] YE X J, LAN S, XIAO S J, et al. Single pole-to-ground fault location method for MMC-HVDC system using wavelet decomposition and DBN[J]. IEEE Transactions on Electrical and Electronic Engineering, 2021, 16(2): 238–247.
[12] 杨茂, 王凯旋. 基于CEEMD-DBN模型的光伏出力日前区间预测[J]. 高电压技术, 2021, 47(4): 1156–1164
YANG Mao, WANG Kaixuan. Day-ahead interval forecasting of PV power based on CEEMD-DBN model[J]. High Voltage Engineering, 2021, 47(4): 1156–1164
[13] 刘冠杰. 高压直流输电线路电晕放电的研究[J]. 山东工业技术, 2018(21): 181–182
LIU Guanjie. Study on corona discharge of HVDC transmission line[J]. Journal of Shandong Industrial Technology, 2018(21): 181–182
[14] 申南轩, 张远航, 徐鹏, 等. 基于气象数据的雾天气条件下高压直流输电线路合成电场计算分析[J]. 中国电机工程学报, 2020, 40(23): 7805–7816
SHEN Nanxuan, ZHANG Yuanhang, XU Peng, et al. Calculation and analysis of ground-level total electric field of HVDC lines in fog based on meteorological data[J]. Proceedings of the CSEE, 2020, 40(23): 7805–7816
[15] KOLEV V G, SULAKOV S I. The weather impact on the overhead line losses[C]//2017 15 th International Conference on Electrical Machines, Drives and Power Systems (ELMA). Sofia, Bulgaria. IEEE, 2017: 119–123.
[16] 陈思佳. 复杂条件下750 kV同塔四回输电线路导线电晕损失评估[D]. 北京: 华北电力大学, 2019: 1–5.
CHEN Sijia. Research on evaluation of corona loss in 750 kV four-circuit transmission lines on the same tower considering complex meteorological conditions[D]. Beijing: North China Electric Power University, 2019: 1–5.
[17] GHOSH S, AHMED N, BANERJEE S. Impact of weather(fog) on corona loss and its geographical variation within eastern region[C]//2018 20 th National Power Systems Conference (NPSC). Tiruchirappalli, India. IEEE, 2018: 1–6.
[18] 邓蓉婕, 方兆本. 基于斯皮尔曼相关分析的理财产品收益率分析[J]. 统计与决策, 2019, 35(16): 164–167
Deng Rongfang, Fang Zhaoben. Analysis of financial product yield based on Spearman correlation analysis[J]. Statistics and Decision, 2019, 35(16): 164–167
[19] 兰文宝, 车畅, 陶成云. 基于斯皮尔曼等级相关的单演谱成分选择及其在SAR目标识别中的应用[J]. 电波科学学报, 2020, 35(3): 414–421
LAN Wenbao, CHE Chang, TAO Chengyun. Selection of monogenic components based on Spearman rank correlation with application to SAR target recognition[J]. Chinese Journal of Radio Science, 2020, 35(3): 414–421
[20] 张新生, 蔡宝泉. 基于改进随机森林模型的海底管道腐蚀预测[J]. 中国安全科学学报, 2021, 31(8): 69–74
ZHANG Xinsheng, CAI Baoquan. Corrosion prediction of submarine pipelines based on improved Random Forest model[J]. China Safety Science Journal, 2021, 31(8): 69–74
[21] 李正明, 梁彩霞, 王满商. 基于PSO-DBN神经网络的光伏短期发电出力预测[J]. 电力系统保护与控制, 2020, 48(8): 149–154
Li Zhengming, Liang Caixia, Wang Manshang. Short-term power generation output prediction based on a PSO-DBN neural network[J]. Power System Protection and Control, 2020, 48(8): 149–154
[22] 叶鑫杰, 兰生, 肖思捷, 等. 基于小波包能量熵和DBN的MMC-HVDC输电线路单极接地故障定位方法[J]. 南方电网技术, 2021, 15(2): 82–91
YE Xinjie, LAN Sheng, XIAO Sijie, et al. Single pole grounding fault location method of MMC-HVDC transmission line based on wavelet packet energy entropy and DBN[J]. Southern Power System Technology, 2021, 15(2): 82–91
[23] 高伟, 杨耿杰, 郭谋发, 等. 基于DTCWT-DBN的配电网内部过电压类型识别[J]. 电力系统保护与控制, 2019, 47(9): 80–89
GAO Wei, YANG Gengjie, GUO Moufa, et al. Internal overvoltage type identification for distribution network based on DTCWT-DBN algorithm[J]. Power System Protection and Control, 2019, 47(9): 80–89
[24] LEE J H, KIM J W, SONG J Y, et al. Distance-based intelligent particle swarm optimization for optimal design of permanent magnet synchronous machine[J]. IEEE Transactions on Magnetics, 2017, 53(6): 1–4.
[25] 李海天. 基于PSO-ESN的线损模拟预测方法研究[D]. 北京: 华北电力大学, 2018.
LI Haitian. Research on line loss rate forecasting using echo state network and optimizing by PSO algorithm[D]. Beijing: North China Electric Power University, 2018.
[26] 沈英, 黎乐谦, 黄峰, 等. 基于粒子群算法的旋转双棱镜指向误差校正方法[J]. 光学学报, 2021, 41(24): 139–148
SHEN Ying, LI Leqian, HUANG Feng, et al. Pointing error correction of risley-prism system based on particle swarm algorithm[J]. Acta Optica Sinica, 2021, 41(24): 139–148
[27] 邢晓溪. 粒子群算法研究进展[J]. 数据通信, 2015(3): 19–21,30
XIN Xiaoxi. Research progress of particle swarm optimization[J]. Data Communication, 2015(3): 19–21,30