计及尖峰电价机制的短期负荷预测研究
|
摘要
针对需求响应的实施对电力系统短期负荷预测带来的新挑战,本文构建了计及尖峰电价需求响应机制的电力系统短期负荷预测模型。研究了基于尖峰电价的需求响应机理,依据基于消费者心理学的用户响应模型,对实施需求响应后的负荷序列进行模拟。同时,构造了Elman神经网络短期负荷预测模型,由于Elman神经网络的承接层反馈使其具有较好的动态存储功能,使模型对非线性负荷序列具有良好的预测性能,并给出实际算例进行仿真分析。仿真结果表明,采用计及尖峰电价需求响应机制的Elman神经网络预测模型,能较准确预测在需求响应策略影响下负荷曲线的变化,最大相对误差为4.34%,平均绝对误差为2.14%;而未考虑需求响应的Elman神经网络预测模型,预测精度明显较低,其最大相对误差为10.76%,平均绝对误差为6.71%,说明将需求响应作为影响因素的预测模型可有效提高模型的预测精度。该研究为计及需求响应的短期负荷预测提供了理论依据。
In view of the new challenges brought by the implementation of demand response to short-term load forecasting of power system,this paper constructs a short-term load forecasting model of power system considering the demand response mechanism of peak electricity price.The demand response mechanism based on peak electricity price is studied,and the load sequence after implementing demand response is simulated according to the user response model based on consumer psychology.At the same time,a short-term load forecasting model based on Elman neural network is constructed.Because of the feedback of the receiving layer of Elman neural network,it has better dynamic storage function,which makes the model have good forecasting performance for the non-linear load series.The simulation analysis is carried out by an actual example.The simulation results show that the Elman neural network forecasting model considering the demand response mechanism of peak electricity price can accurately predict the change of load curve under the influence of demand response strategy,with the maximum relative error of 4.34% and the average absolute error of 2.14%.The Elman neural network forecasting model without considering demand response has obviously lower forecasting accuracy,with the maximum relative error of 10.76%.The average absolute error is 6.71%,which shows that the prediction model with demand response as the influencing factor can effectively improve the prediction accuracy of the model.This study provides a theoretical basis for short-term load forecasting considering demand response.
In view of the new challenges brought by the implementation of demand response to short-term load forecasting of power system,this paper constructs a short-term load forecasting model of power system considering the demand response mechanism of peak electricity price.The demand response mechanism based on peak electricity price is studied,and the load sequence after implementing demand response is simulated according to the user response model based on consumer psychology.At the same time,a short-term load forecasting model based on Elman neural network is constructed.Because of the feedback of the receiving layer of Elman neural network,it has better dynamic storage function,which makes the model have good forecasting performance for the non-linear load series.The simulation analysis is carried out by an actual example.The simulation results show that the Elman neural network forecasting model considering the demand response mechanism of peak electricity price can accurately predict the change of load curve under the influence of demand response strategy,with the maximum relative error of 4.34% and the average absolute error of 2.14%.The Elman neural network forecasting model without considering demand response has obviously lower forecasting accuracy,with the maximum relative error of 10.76%.The average absolute error is 6.71%,which shows that the prediction model with demand response as the influencing factor can effectively improve the prediction accuracy of the model.This study provides a theoretical basis for short-term load forecasting considering demand response.
引文
[1]陆苏青,唐楠,王蓓蓓,等.美国需求响应技术和思考(下)[J].电力需求侧管理,2016,17(1):55-61.
[2] Sharifi R,Anvari-Moghaddam A,Fathi S H,et al.Economic demand response model in liberalisedelectricity markets with respect to flexibility of consumers[J].Iet Generation Transmission&Distribution,2017,11(17):4291-4298.
[3]李娜,张文月,王玉玮,等.基于数据均值化及LSSVM算法的峰谷电价需求响应模型[J].中国电力,2016,49(9):137-141.
[4]孙宇军,王岩,王蓓蓓,等.考虑需求响应不确定性的多时间尺度源荷互动决策方法[J].电力系统自动化,2018,42(2):106-113.
[5]梁甜甜,王磊,高赐威.基于用户响应的尖峰电价模型研究[J].华东电力,2013,41(1):42-46.
[6]马永武,赵国生,黄明山,等.峰谷分时电价下用户需求响应行为模型的研究[J].郑州大学学报:理学版,2015,47(4):119-122.
[7]刘荣,方鸽飞.改进Elman神经网络的综合气象短期负荷预测[J].电力系统保护与控制,2012,40(22):113-117.
[8]陈艳平,毛弋,陈萍,等.基于EEMD-样本熵和Elman神经网络的短期电力负荷预测[J].电力系统及其自动化学报,2016,28(3):59-64.
[9]杨丽君,李健强,李学平,等.考虑需求响应的含风电电力系统日前经济调度[J].电工电能新技术,2015,34(11):29-36.
[10]王蓓蓓.面向智能电网的用户需求响应特性和能力研究综述[J].中国电机工程学报,2014,34(22):3654-3663.
[11]张舒菡.智能电网条件下的需求响应关键技术[J].电子技术与软件工程,2017,34(6):3576-3589.
[12]苏卫华,储琳琳,张亮,等.考虑需求侧管理的负荷预测方法研究[J].华东电力,2010,38(8):1236-1239.
[13] Moghaddam M P,Abdollahi A,Rashidinejad M.Flexible demand response programs modeling in competitive electricity markets[J].Applied Energy,2011,88(9):3257-3269.
[14]高赐威,陈曦寒,陈江华,等.我国电力需求响应的措施与应用方法[J].电力需求侧管理,2013,15(1):1-4,6.
[15]朱晟,蒋传文,侯志俭.基于气象负荷因子的Elman神经网络短期负荷预测[J].电力系统及其自动化学报,2005,17(1):23-26.
[16] Sharifi R,Anvari-Moghaddam A,Fathi S H,et al.An economic customer-oriented demand response model in electricity markets[C]∥IEEE International Conference on Industrial Technology-Icit.Lyon,France:IEEE,2018.
[17] Kelo S,Dudul S.A wavelet elmanneural network for short-term electrical load prediction under the influence of temperature[J].International Journal of Electrical Power&Energy Systems,2012,43(1):1063-1071.
[18] Albadi M H,El-Saadany E F.A summary of demand response in electricity markets[J].Electric Power Systems Research,2008,78(11):1989-1996.
[19] Herter K,Mcauliffe P,Rosenfeld A.An exploratory analysis of californiaresidential customer response to critical peak pricing of electricity[J].Energy,2007,32(1):25-34.
[20] Zhang Q,Wang X F,Fu M.Optimal implementation strategies for critical peak pricing[C]∥International Conference on the European Energy Market.Leuven,Belgium:IEEE,2009:27-29.
[2] Sharifi R,Anvari-Moghaddam A,Fathi S H,et al.Economic demand response model in liberalisedelectricity markets with respect to flexibility of consumers[J].Iet Generation Transmission&Distribution,2017,11(17):4291-4298.
[3]李娜,张文月,王玉玮,等.基于数据均值化及LSSVM算法的峰谷电价需求响应模型[J].中国电力,2016,49(9):137-141.
[4]孙宇军,王岩,王蓓蓓,等.考虑需求响应不确定性的多时间尺度源荷互动决策方法[J].电力系统自动化,2018,42(2):106-113.
[5]梁甜甜,王磊,高赐威.基于用户响应的尖峰电价模型研究[J].华东电力,2013,41(1):42-46.
[6]马永武,赵国生,黄明山,等.峰谷分时电价下用户需求响应行为模型的研究[J].郑州大学学报:理学版,2015,47(4):119-122.
[7]刘荣,方鸽飞.改进Elman神经网络的综合气象短期负荷预测[J].电力系统保护与控制,2012,40(22):113-117.
[8]陈艳平,毛弋,陈萍,等.基于EEMD-样本熵和Elman神经网络的短期电力负荷预测[J].电力系统及其自动化学报,2016,28(3):59-64.
[9]杨丽君,李健强,李学平,等.考虑需求响应的含风电电力系统日前经济调度[J].电工电能新技术,2015,34(11):29-36.
[10]王蓓蓓.面向智能电网的用户需求响应特性和能力研究综述[J].中国电机工程学报,2014,34(22):3654-3663.
[11]张舒菡.智能电网条件下的需求响应关键技术[J].电子技术与软件工程,2017,34(6):3576-3589.
[12]苏卫华,储琳琳,张亮,等.考虑需求侧管理的负荷预测方法研究[J].华东电力,2010,38(8):1236-1239.
[13] Moghaddam M P,Abdollahi A,Rashidinejad M.Flexible demand response programs modeling in competitive electricity markets[J].Applied Energy,2011,88(9):3257-3269.
[14]高赐威,陈曦寒,陈江华,等.我国电力需求响应的措施与应用方法[J].电力需求侧管理,2013,15(1):1-4,6.
[15]朱晟,蒋传文,侯志俭.基于气象负荷因子的Elman神经网络短期负荷预测[J].电力系统及其自动化学报,2005,17(1):23-26.
[16] Sharifi R,Anvari-Moghaddam A,Fathi S H,et al.An economic customer-oriented demand response model in electricity markets[C]∥IEEE International Conference on Industrial Technology-Icit.Lyon,France:IEEE,2018.
[17] Kelo S,Dudul S.A wavelet elmanneural network for short-term electrical load prediction under the influence of temperature[J].International Journal of Electrical Power&Energy Systems,2012,43(1):1063-1071.
[18] Albadi M H,El-Saadany E F.A summary of demand response in electricity markets[J].Electric Power Systems Research,2008,78(11):1989-1996.
[19] Herter K,Mcauliffe P,Rosenfeld A.An exploratory analysis of californiaresidential customer response to critical peak pricing of electricity[J].Energy,2007,32(1):25-34.
[20] Zhang Q,Wang X F,Fu M.Optimal implementation strategies for critical peak pricing[C]∥International Conference on the European Energy Market.Leuven,Belgium:IEEE,2009:27-29.