基于总体测辨和人工神经网络的负荷建模及预测方法
|
摘要
为提高短期负荷预测精度,提出一种基于总体测辨和人工神经网络的负荷建模及预测方法。通过总体测辨对电力负荷的影响因素进行筛选,得到电力负荷的主要影响因素;然后根据筛选得到的主要影响因素作为输入量建立基于径向基函数神经网络的负荷预测模型,并通过仿真对预测模型进行了验证。仿真结果表明,相比于不采用总体测辨进行影响因素筛选的负荷预测方法,本文方法的平均预测精度提高了2.5%,从而能够有效提高负荷预测精度。
To improve the accuracy of short-term load forecasting,a load modelingand forecasting method is proposed based on the measurement-based method and artificial neural network(ANN). The influencing factors for electric load are screened using the measurement-based method,thus the main influencing factors are obtained;afterwards,these factors are further taken as input to construct a load forecasting model based on the radial basis function(RBF)neural network. This model is verified through simulations,and simulation results show that compared with the load forecasting method that does not use the measurement-based method to screen influencing factors,the average forecasting precision using the proposed method is improved by 2.5%,indicating that the novel method can effectively improve the load forecasting accuracy.
To improve the accuracy of short-term load forecasting,a load modelingand forecasting method is proposed based on the measurement-based method and artificial neural network(ANN). The influencing factors for electric load are screened using the measurement-based method,thus the main influencing factors are obtained;afterwards,these factors are further taken as input to construct a load forecasting model based on the radial basis function(RBF)neural network. This model is verified through simulations,and simulation results show that compared with the load forecasting method that does not use the measurement-based method to screen influencing factors,the average forecasting precision using the proposed method is improved by 2.5%,indicating that the novel method can effectively improve the load forecasting accuracy.
引文
[1]康重庆,夏清,张伯明(Kang Chongqi,Xia Qing,ZhangBoming).电力系统负荷预测研究综述与发展方向的探讨(Review of power system load forecasting and its devel-opment)[J].电力系统自动化(Automation of ElectricPower Systems),2004,28(17):1-11.
[2]龙勇,苏振宇,盖晓平(Long Yong,Su Zhenyu,Gai Xiao-ping).成分分解方法预测月度电力负荷(Monthly loadforecasting using component decomposition method)[J].电力系统及其自动化学报(Proceedings of the CSU-EP-SA),2017,29(5):35-40.
[3]肖白,蒲睿,穆钢,等(Xiao Bai,Pu Rui,Mu Gang,et al).基于多尺度空间分辨率的空间负荷预测误差评价方法(Method of spatial load forecasting error evaluation basedon the multi-scale spatial resolution)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(22):5731-5739.
[4]廖旎焕,胡智宏,马莹莹,等(Liao Nihuan,Hu Zhihong,Ma Yingying,et al).电力系统短期负荷预测方法综述(Review of the short-term load forecasting methods ofelectric power system)[J].电力系统保护与控制(PowerSystem Protection and Control),2011,39(1):147-152.
[5]肖白,周潮,穆钢(Xiao Bai,Zhou Chao,Mu Gang).空间电力负荷预测方法综述与展望(Review and prospect ofthe spatial load forecasting methods)[J].中国电机工程学报(Proceedings of the CSEE),2013,33(25):78-92.
[6]王惠中,周佳,刘轲(Wang Huizhong,Zhou Jia,Liu Ke).电力系统短期负荷预测方法研究综述(Summary of re-search on the short-term load forecasting method of theelectric power system)[J].电气自动化(Electrical Auto-mation),2015,37(1):1-3,39.
[7]李钷,李敏,刘涤尘(Li Po,Li Min,Liu Dichen).基于改进回归法的电力负荷预测(Power load forecasting basedon improved regression)[J].电网技术(Power SystemTechnology),2006,30(1):99-104.
[8]张锋,吴劲晖,张怡,等(Zhang Feng,Wu Jinhui,ZhangYi,et al).基于负荷趋势的新型超短期负荷预测法(Anovel ultra-short term load forecasting based on loadtrend)[J].电网技术(Power System Technology),2004,28(19):64-67.
[9]葛少云,贾鸥莎,刘洪(Ge Shaoyun,Jia Ousha,LiuHong).基于遗传灰色神经网络模型的实时电价条件下短期电力负荷预测(A gray neural network model im-proved by genetic algorithm for short-term load forecast-ing in price-sensitive environment)[J].电网技术(PowerSystem Technology),2012,36(1):224-229.
[10]祝燕萍,方鸽飞(Zhu Yanping,Fang Gefei).基于动态自适应神经网络和人体舒适度的短期负荷预测(Short-term load forecasting based on dynamic adaptive artificialneural network and human body amenity indicator)[J].电力系统保护与控制(Power System Protection and Con-trol),2012,40(1):56-61.
[11]Chen Ying,Luh P B,Guan Che,et al.Short-term load fore-casting:Similar day-based wavelet neural networks[J].IEEE Trans on Power Systems,2010,25(1):322-330.
[12]Zhang Yun,Zhou Quan,Sun Caixin,et al.RBF neural net-work and ANFIS-based short-term load forecasting ap-proach in real-time price environment[J].IEEE Trans onPower Systems,2008,23(3):853-858.
[13]孟洋洋,卢继平,孙华利,等(Meng Yangyang,Lu Jiping,Sun Huali,et al).基于相似日和人工神经网络的风电功率短期预测(Short-term wind power forecasting basedon similar days and artificial neural network)[J].电网技术(Power System Technology),2010,34(12):163-167.
[14]王宁,谢敏,邓佳梁,等(Wang Ning,Xie Min,Deng Jia-liang,et al).基于支持向量机回归组合模型的中长期降温负荷预测(Mid-long term temperature-lowering loadforecasting based on combination of support vector machineand multiple regression)[J].电力系统保护与控制(Pow-er System Protection and Control),2016,44(3):92-97.
[15]王德意,杨卓,杨国清(Wang Deyi,Yang Zhuo,YangGuoqing).基于负荷混沌特性和最小二乘支持向量机的短期负荷预测(Short-term load forecasting based onchaotic characteristic of loads and least squares supportvector machines)[J].电网技术(Power System Technolo-gy),2008,32(7):66-71.
[16]李霄,王昕,郑益慧,等(Li Xiao,Wang Xin,Zheng Yihui,et al).基于改进最小二乘支持向量机和预测误差校正的短期风电负荷预测(Short-term wind load forecastingbased on improved LSSVM and error forecasting correc-tion)[J].电力系统保护与控制(Power System Protectionand Control),2015,43(11):63-69.
[17]Ceperic E,Ceperic V,Baric A.A strategy for short-termload forecasting by support vector regression machines[J].IEEE Trans on Power Systems,2013,28(4):4356-4364.
[18]Liu Bin,Su Hongye,Huang Weihua,et al.Temperatureprediction control based on least squares support vectormachines[J].Journal of Control Theory and Applications,2004,2(4):365-370.
[19]潘峰,程浩忠,杨镜非,等(Pan Feng,Cheng Haozhong,Yang Jingfei,et al).基于支持向量机的电力系统短期负荷预测(Power system short-term load forecasting basedon support vector machines)[J].电网技术(Power SystemTechnology),2004,28(21):39-42.
[20]姚李孝,刘学琴(Yao Lixiao,Liu Xueqin).基于小波分析的月度负荷组合预测(A wavelet analysis based com-bined model for monthly load forecasting)[J].电网技术(Power System Technology),2007,31(19):65-68.
[21]邰能灵,侯志俭,李涛,等(Tai Nengling,Hou Zhijian,LiTao,et al).基于小波分析的电力系统短期负荷预测方法(New principle based on wavelet transform for powersystem short-term load forecasting)[J].中国电机工程学报(Proceedings of the CSEE),2003,23(1):45-50.
[22]张红斌,汤涌,张东霞,等(Zhang Hongbin,Tang Yong,Zhang Dongxia,et al).基于总体测辨法的电力负荷建模系统(Load modeling system founded on measurement-based method)[J].电网技术(Power System Technology),2007,31(4):32-35.
[23]李欣然,惠金花,钱军,等(Li Xinran,Hui Jinhua,QianJun,et al).风力发电对配电网侧负荷建模的影响(Im-pact of wind power generation on load modeling in distri-bution network)[J].电力系统自动化(Automation ofElectric Power Systems),2009,33(13):89-94.
[24]包宇庆,李扬,杨斌,等(Bao Yuqing,Li Yang,Yang Bin,et al).基于神经网络的日峰荷预测方法中日期类型系数的确定(Weekday index determination for ANN-baseddaily peak load forecasting method)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(22):5715-5722.
[2]龙勇,苏振宇,盖晓平(Long Yong,Su Zhenyu,Gai Xiao-ping).成分分解方法预测月度电力负荷(Monthly loadforecasting using component decomposition method)[J].电力系统及其自动化学报(Proceedings of the CSU-EP-SA),2017,29(5):35-40.
[3]肖白,蒲睿,穆钢,等(Xiao Bai,Pu Rui,Mu Gang,et al).基于多尺度空间分辨率的空间负荷预测误差评价方法(Method of spatial load forecasting error evaluation basedon the multi-scale spatial resolution)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(22):5731-5739.
[4]廖旎焕,胡智宏,马莹莹,等(Liao Nihuan,Hu Zhihong,Ma Yingying,et al).电力系统短期负荷预测方法综述(Review of the short-term load forecasting methods ofelectric power system)[J].电力系统保护与控制(PowerSystem Protection and Control),2011,39(1):147-152.
[5]肖白,周潮,穆钢(Xiao Bai,Zhou Chao,Mu Gang).空间电力负荷预测方法综述与展望(Review and prospect ofthe spatial load forecasting methods)[J].中国电机工程学报(Proceedings of the CSEE),2013,33(25):78-92.
[6]王惠中,周佳,刘轲(Wang Huizhong,Zhou Jia,Liu Ke).电力系统短期负荷预测方法研究综述(Summary of re-search on the short-term load forecasting method of theelectric power system)[J].电气自动化(Electrical Auto-mation),2015,37(1):1-3,39.
[7]李钷,李敏,刘涤尘(Li Po,Li Min,Liu Dichen).基于改进回归法的电力负荷预测(Power load forecasting basedon improved regression)[J].电网技术(Power SystemTechnology),2006,30(1):99-104.
[8]张锋,吴劲晖,张怡,等(Zhang Feng,Wu Jinhui,ZhangYi,et al).基于负荷趋势的新型超短期负荷预测法(Anovel ultra-short term load forecasting based on loadtrend)[J].电网技术(Power System Technology),2004,28(19):64-67.
[9]葛少云,贾鸥莎,刘洪(Ge Shaoyun,Jia Ousha,LiuHong).基于遗传灰色神经网络模型的实时电价条件下短期电力负荷预测(A gray neural network model im-proved by genetic algorithm for short-term load forecast-ing in price-sensitive environment)[J].电网技术(PowerSystem Technology),2012,36(1):224-229.
[10]祝燕萍,方鸽飞(Zhu Yanping,Fang Gefei).基于动态自适应神经网络和人体舒适度的短期负荷预测(Short-term load forecasting based on dynamic adaptive artificialneural network and human body amenity indicator)[J].电力系统保护与控制(Power System Protection and Con-trol),2012,40(1):56-61.
[11]Chen Ying,Luh P B,Guan Che,et al.Short-term load fore-casting:Similar day-based wavelet neural networks[J].IEEE Trans on Power Systems,2010,25(1):322-330.
[12]Zhang Yun,Zhou Quan,Sun Caixin,et al.RBF neural net-work and ANFIS-based short-term load forecasting ap-proach in real-time price environment[J].IEEE Trans onPower Systems,2008,23(3):853-858.
[13]孟洋洋,卢继平,孙华利,等(Meng Yangyang,Lu Jiping,Sun Huali,et al).基于相似日和人工神经网络的风电功率短期预测(Short-term wind power forecasting basedon similar days and artificial neural network)[J].电网技术(Power System Technology),2010,34(12):163-167.
[14]王宁,谢敏,邓佳梁,等(Wang Ning,Xie Min,Deng Jia-liang,et al).基于支持向量机回归组合模型的中长期降温负荷预测(Mid-long term temperature-lowering loadforecasting based on combination of support vector machineand multiple regression)[J].电力系统保护与控制(Pow-er System Protection and Control),2016,44(3):92-97.
[15]王德意,杨卓,杨国清(Wang Deyi,Yang Zhuo,YangGuoqing).基于负荷混沌特性和最小二乘支持向量机的短期负荷预测(Short-term load forecasting based onchaotic characteristic of loads and least squares supportvector machines)[J].电网技术(Power System Technolo-gy),2008,32(7):66-71.
[16]李霄,王昕,郑益慧,等(Li Xiao,Wang Xin,Zheng Yihui,et al).基于改进最小二乘支持向量机和预测误差校正的短期风电负荷预测(Short-term wind load forecastingbased on improved LSSVM and error forecasting correc-tion)[J].电力系统保护与控制(Power System Protectionand Control),2015,43(11):63-69.
[17]Ceperic E,Ceperic V,Baric A.A strategy for short-termload forecasting by support vector regression machines[J].IEEE Trans on Power Systems,2013,28(4):4356-4364.
[18]Liu Bin,Su Hongye,Huang Weihua,et al.Temperatureprediction control based on least squares support vectormachines[J].Journal of Control Theory and Applications,2004,2(4):365-370.
[19]潘峰,程浩忠,杨镜非,等(Pan Feng,Cheng Haozhong,Yang Jingfei,et al).基于支持向量机的电力系统短期负荷预测(Power system short-term load forecasting basedon support vector machines)[J].电网技术(Power SystemTechnology),2004,28(21):39-42.
[20]姚李孝,刘学琴(Yao Lixiao,Liu Xueqin).基于小波分析的月度负荷组合预测(A wavelet analysis based com-bined model for monthly load forecasting)[J].电网技术(Power System Technology),2007,31(19):65-68.
[21]邰能灵,侯志俭,李涛,等(Tai Nengling,Hou Zhijian,LiTao,et al).基于小波分析的电力系统短期负荷预测方法(New principle based on wavelet transform for powersystem short-term load forecasting)[J].中国电机工程学报(Proceedings of the CSEE),2003,23(1):45-50.
[22]张红斌,汤涌,张东霞,等(Zhang Hongbin,Tang Yong,Zhang Dongxia,et al).基于总体测辨法的电力负荷建模系统(Load modeling system founded on measurement-based method)[J].电网技术(Power System Technology),2007,31(4):32-35.
[23]李欣然,惠金花,钱军,等(Li Xinran,Hui Jinhua,QianJun,et al).风力发电对配电网侧负荷建模的影响(Im-pact of wind power generation on load modeling in distri-bution network)[J].电力系统自动化(Automation ofElectric Power Systems),2009,33(13):89-94.
[24]包宇庆,李扬,杨斌,等(Bao Yuqing,Li Yang,Yang Bin,et al).基于神经网络的日峰荷预测方法中日期类型系数的确定(Weekday index determination for ANN-baseddaily peak load forecasting method)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(22):5715-5722.