基于压缩感知理论的缺失数据集下线损预测模型
|
摘要
线损预测是电网企业进行线损管理的基础,而电力系统中数据收集与传输过程中不可避免出现各种异常状况,导致线损数据缺失,影响线损预测精度。为解决这一问题,应用压缩感知理论研究矩阵稀疏变换方法和矩阵重构算法,实现电网运行缺失数据的补全与重建,利用基于自适应噪声的完整集成经验模态分解建立线损预测模型,完成缺失数据集下的线损预测。某10 kV配电网算例验证结果表明,在数据量较大或数据缺失情况较严重的情况下,基于压缩感知理论的数据恢复方法能比传统方法更好地修复原始数据,恢复原始数据的变化趋势,提高线损预测精度。
Line loss prediction is the basis for line loss management by power grid enterprises, while avoidable abnormal conditions in data collection and transmission in the power system may cause line loss data missing and affect prediction precision. To solve the problem, this paper applies the compressed sensing theory in studying the method of matrix sparse transformation and matrix reconstruction so as to finish completion and reconstruction of missing data. It uses complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) to establish the line loss prediction model and finish line loss prediction under missing data set. An actual example of one 10 kV power distribution network has verified that under the condition of large data amount or serious data missing, the data repairing method based on the compressed sensing theory is preferable to repair original data, recover variation trend of original data and improve line loss prediction precision compared with traditional methods
Line loss prediction is the basis for line loss management by power grid enterprises, while avoidable abnormal conditions in data collection and transmission in the power system may cause line loss data missing and affect prediction precision. To solve the problem, this paper applies the compressed sensing theory in studying the method of matrix sparse transformation and matrix reconstruction so as to finish completion and reconstruction of missing data. It uses complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) to establish the line loss prediction model and finish line loss prediction under missing data set. An actual example of one 10 kV power distribution network has verified that under the condition of large data amount or serious data missing, the data repairing method based on the compressed sensing theory is preferable to repair original data, recover variation trend of original data and improve line loss prediction precision compared with traditional methods
引文
[1] 施寅跃,付艳兰.地区电网线损过程管理评价体系研究[J].云南电力技术,2018,46(5):46-48. SHI Yinyue, FU Yanlan. Research and establish the evaluation system of line loss process management[J]. Yunnan Electric Power, 2018, 46(5):46-48.
[2] 彭剑宁.基于可拓层次分析法的县级电网线损管理评价[J].云南电力技术,2018,46(3):29-31. PENG Jianning. Evaluate on line-loss-management of county-power-grid Based on the extension of analytic hierarchy process[J]. Yunnan Electric Power, 2018,46(3):29-31.
[3] 张永凯,马欢,张玮,等.计及峰谷分时的配电线路同期月线损预测方法及应用[J].电网与清洁能源,2017,33(10):67-73. ZHANG Yongkai, MA Huan, ZHANG Wei, et al. Forcasting method of correspongding period monthly line loss in a distribuion network and its application considering time of use[J]. Power System and Clean Energy, 2017,33(10):67-73.
[4] 牛瑞,张望妮.SCADA系统异常数据分析及治理建议[J].电网与清洁能源,2017,33(9):68-71. NIU Rui,ZHANG Wangni. Abnormal data analysis and treatment recommendations for SCADA System[J]. Power System and Clean Energy, 2017,33(9):68-71.
[5] 宋亚奇,周国亮,朱永利.智能电网大数据处理技术现状与挑战[J].电网技术,2013,37(4):927-935. SONG Yaqi, ZHOU Guoliang, ZHU Yongli. Present status and challenges of big data processing in smart grid[J]. Power System Technology, 2013,37(4):927-935.
[6] 邹云峰,梅飞,李悦,等.基于数据挖掘技术的台区合理线损预测模型研究[J].电力需求侧管理,2015(4):25-29. ZOU Yunfeng, MEI Fei, LI Yue, et al. Prediction model research of reasonable line loss for transformer district based on data mining technology[J]. Power Demand Side Management, 2015(4):25-29.
[7] 黄文思,许元斌,邹保平,等. 基于大数据的线损计算分析研究[J].电气应用,2015(20):79-84. HUANG Wensi,XU Yuanbing,ZOU Baoping,et al. Analysis of line loss based on large data[J]. Electrotechnical Application, 2015(20):79-84.
[8] 胡可为,王蕾,董运昌,等. 基于Spark的电力生产数据挖掘平台设计[ J ]. 电力大数据,2017,20(9):20-22. HU Kewei,WANG Lei,DONG Yunchang,et al. Design of power production data mining platform based on Spark[J]. Power systems and big data,2017, 20(9):20-22.
[9] 卢建昌,樊围国.大数据时代下数据挖掘技术在电力企业中的应用[J].广东电力,2014,27(9):88-94. LU Jianchang, FAN Weiguo. Application of data mining technology in electric power enterprises in era of big data[J]. Guangdong Electric Power, 2014,27(9):88-94.
[10] 陈雷,郑德忠,王忠东. 基于压缩感知的电力信号压缩与重构研究[J].电测与仪表,2017,54(1):95-100. CHEN Lei, ZHENG Dezhong, WANG Zhongdong. Research on power signal compression and reconstruction based on compressed sensing[J]. Electrical Measurement & Instrumentation, 2017,54(1):95-100.
[11] 上官云祺. 基于压缩感知的电力系统数据采集与重构算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[12] DONOHO D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
[13] 李军,李青. 基于CEEMDAN-排列熵和泄漏积分ESN的中期电力负荷预测研究[J]. 电机与控制学报,2015(8):70-80. LI Jun, LI Qing. Medium term electricity load forecasting based on CEEMDAN-permutation entropy and ESN with leaky integrator neurons[J]. Electric Machines and Control, 2015(8):70-80.
[14] TORRES M E, COLOMINAS M A, SCHLOTTHAUER G, et al. A complete ensemble empirical mode decomposition with adaptive noise[J]. Brain Research Bulletin, 2011, 125(3):4144-4147.
[15] 李锋,林阳阳,赵辉,等. 基于CEEMDAN-SVM的液压泵故障诊断方法研究[J]. 液压与气动,2016(1):125-129. LI Feng,LIN Yangyang,ZHAO Hui,et al. Fault diagnosis of hydraulic pump based on CEEMDAN-SVM[J]. Chinese Hydraulics & Pneumatics, 2016(1):125-129.
[16] 张新鹏,胡茑庆,程哲,等. 基于压缩感知的振动数据修复方法[J]. 物理学报,2014(20):119-128. ZHANG Xinpeng,HU Niaoqing,CHENG Zhe,et al. Vibration data recovery based on compressed sensing[J]. Acta Physica Sinica, 2014(20):119-128.
[17] 李珅,马彩文,李艳,等. 压缩感知重构算法综述[J].红外与激光工程,2013,42(增刊1):225-232. LI Shen, MA Caiwen, LI Yan, et al. Survey on reconstruction algorithm based on compressive sensing[J]. Infrared and Laser Engineering, 2013,42(S1):225-232.
[18] 徐立军. 压缩感知重构算法及其应用研究[D].太原:中北大学,2016.
[19] 王大为. 数据压缩方法研究及其在电力系统中的应用[D].秦皇岛:燕山大学,2014.
[20] 茆美琴,龚文剑,张榴晨,等. 基于EEMD-SVM方法的光伏电站短期出力预测[J]. 中国电机工程学报,2013(34):17-24,5. MAO Meiqin, GONG Wenjian, ZHANG Liuchen, et al. Short-term photovoltaic generation forecasting based on EEMD-SVM combined method[J]. Proceedings of the CSEE, 2013(34):17-24,5.
[21] 尹东阳,盛义发,蒋明洁,等. 基于粗糙集理论-主成分分析的Elman神经网络短期风速预测[J]. 电力系统保护与控制,2014(11):46-51. YIN Dongyang, SHENG Yifa, JIANG Mingjie, et al. Short-term wind speed forecasting using Elman neural network based on rough set theory and principal components analysis[J]. Power System Protection and Control, 2014(11):46-51.
[2] 彭剑宁.基于可拓层次分析法的县级电网线损管理评价[J].云南电力技术,2018,46(3):29-31. PENG Jianning. Evaluate on line-loss-management of county-power-grid Based on the extension of analytic hierarchy process[J]. Yunnan Electric Power, 2018,46(3):29-31.
[3] 张永凯,马欢,张玮,等.计及峰谷分时的配电线路同期月线损预测方法及应用[J].电网与清洁能源,2017,33(10):67-73. ZHANG Yongkai, MA Huan, ZHANG Wei, et al. Forcasting method of correspongding period monthly line loss in a distribuion network and its application considering time of use[J]. Power System and Clean Energy, 2017,33(10):67-73.
[4] 牛瑞,张望妮.SCADA系统异常数据分析及治理建议[J].电网与清洁能源,2017,33(9):68-71. NIU Rui,ZHANG Wangni. Abnormal data analysis and treatment recommendations for SCADA System[J]. Power System and Clean Energy, 2017,33(9):68-71.
[5] 宋亚奇,周国亮,朱永利.智能电网大数据处理技术现状与挑战[J].电网技术,2013,37(4):927-935. SONG Yaqi, ZHOU Guoliang, ZHU Yongli. Present status and challenges of big data processing in smart grid[J]. Power System Technology, 2013,37(4):927-935.
[6] 邹云峰,梅飞,李悦,等.基于数据挖掘技术的台区合理线损预测模型研究[J].电力需求侧管理,2015(4):25-29. ZOU Yunfeng, MEI Fei, LI Yue, et al. Prediction model research of reasonable line loss for transformer district based on data mining technology[J]. Power Demand Side Management, 2015(4):25-29.
[7] 黄文思,许元斌,邹保平,等. 基于大数据的线损计算分析研究[J].电气应用,2015(20):79-84. HUANG Wensi,XU Yuanbing,ZOU Baoping,et al. Analysis of line loss based on large data[J]. Electrotechnical Application, 2015(20):79-84.
[8] 胡可为,王蕾,董运昌,等. 基于Spark的电力生产数据挖掘平台设计[ J ]. 电力大数据,2017,20(9):20-22. HU Kewei,WANG Lei,DONG Yunchang,et al. Design of power production data mining platform based on Spark[J]. Power systems and big data,2017, 20(9):20-22.
[9] 卢建昌,樊围国.大数据时代下数据挖掘技术在电力企业中的应用[J].广东电力,2014,27(9):88-94. LU Jianchang, FAN Weiguo. Application of data mining technology in electric power enterprises in era of big data[J]. Guangdong Electric Power, 2014,27(9):88-94.
[10] 陈雷,郑德忠,王忠东. 基于压缩感知的电力信号压缩与重构研究[J].电测与仪表,2017,54(1):95-100. CHEN Lei, ZHENG Dezhong, WANG Zhongdong. Research on power signal compression and reconstruction based on compressed sensing[J]. Electrical Measurement & Instrumentation, 2017,54(1):95-100.
[11] 上官云祺. 基于压缩感知的电力系统数据采集与重构算法研究[D].哈尔滨:哈尔滨工业大学,2014.
[12] DONOHO D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
[13] 李军,李青. 基于CEEMDAN-排列熵和泄漏积分ESN的中期电力负荷预测研究[J]. 电机与控制学报,2015(8):70-80. LI Jun, LI Qing. Medium term electricity load forecasting based on CEEMDAN-permutation entropy and ESN with leaky integrator neurons[J]. Electric Machines and Control, 2015(8):70-80.
[14] TORRES M E, COLOMINAS M A, SCHLOTTHAUER G, et al. A complete ensemble empirical mode decomposition with adaptive noise[J]. Brain Research Bulletin, 2011, 125(3):4144-4147.
[15] 李锋,林阳阳,赵辉,等. 基于CEEMDAN-SVM的液压泵故障诊断方法研究[J]. 液压与气动,2016(1):125-129. LI Feng,LIN Yangyang,ZHAO Hui,et al. Fault diagnosis of hydraulic pump based on CEEMDAN-SVM[J]. Chinese Hydraulics & Pneumatics, 2016(1):125-129.
[16] 张新鹏,胡茑庆,程哲,等. 基于压缩感知的振动数据修复方法[J]. 物理学报,2014(20):119-128. ZHANG Xinpeng,HU Niaoqing,CHENG Zhe,et al. Vibration data recovery based on compressed sensing[J]. Acta Physica Sinica, 2014(20):119-128.
[17] 李珅,马彩文,李艳,等. 压缩感知重构算法综述[J].红外与激光工程,2013,42(增刊1):225-232. LI Shen, MA Caiwen, LI Yan, et al. Survey on reconstruction algorithm based on compressive sensing[J]. Infrared and Laser Engineering, 2013,42(S1):225-232.
[18] 徐立军. 压缩感知重构算法及其应用研究[D].太原:中北大学,2016.
[19] 王大为. 数据压缩方法研究及其在电力系统中的应用[D].秦皇岛:燕山大学,2014.
[20] 茆美琴,龚文剑,张榴晨,等. 基于EEMD-SVM方法的光伏电站短期出力预测[J]. 中国电机工程学报,2013(34):17-24,5. MAO Meiqin, GONG Wenjian, ZHANG Liuchen, et al. Short-term photovoltaic generation forecasting based on EEMD-SVM combined method[J]. Proceedings of the CSEE, 2013(34):17-24,5.
[21] 尹东阳,盛义发,蒋明洁,等. 基于粗糙集理论-主成分分析的Elman神经网络短期风速预测[J]. 电力系统保护与控制,2014(11):46-51. YIN Dongyang, SHENG Yifa, JIANG Mingjie, et al. Short-term wind speed forecasting using Elman neural network based on rough set theory and principal components analysis[J]. Power System Protection and Control, 2014(11):46-51.