基于SA-BP算法的蒙西电力现货市场价格预测Price Forecasting for West Inner Mongolia Power Spot Market Based on SA-BP Algorithm
王金鑫,蒋子彦,王琪,李东,于福荣,刘海,任婧媛
WANG Jinxin,JIANG Ziyan,WANG Qi,LI Dong,YU Furong,LIU Hai,REN Jingyuan
摘要(Abstract):
针对传统BP神经网络易陷入局部最优的问题,结合模拟退火算法中的随机扰动机制和BP神经网络的梯度下降法对神经网络阈值与权值进行调整,从而避免BP网络陷入局部最优陷阱。同时针对蒙西电力市场外送电量、新能源装机占比高的特点,将样本预处理为预测净负荷与历史电价,以减少因外送电量与新能源因素产生的预测误差。最后采用蒙西电力市场历史运行数据对电价进行预测。结果表明,经模拟退火算法优化后的BP神经网络模型可根据蒙西电力市场特征,对电价进行有效预测。
The traditional BP neural network is easy to fall into the local optimal solution. To deal with this problem, the threshold and weight of the BP neural network are adjusted by combining the random disturbance in the simulated annealing algorithm. Considering the characteristics of high proportion of external power supply and renewable energy installation in West Inner Mongolia power market, the samples are preprocessed to predict net load and historical electricity price to minimize the predict error. The historical data of a certain period of time in West Inner Mongolia power market are used to predict the electricity price. The result shows that BP neural network model optimized by simulated annealing algorithm can effectively predict the electricity price after considering the market characteristics of West Inner Mongolia power market.
关键词(KeyWords):
模拟退火算法;BP神经网络算法;电力现货市场;电价预测;局部最优
simulated annealing algorithm;BP neural network algorithm;power spot market;electricity price forecasting;local optimal solution
基金项目(Foundation): 北京京能电力股份有限公司科技项目“面向蒙西电力现货市场的辅助决策系统研究及京能电力在蒙西现货市场的运营效益分析”(DL202001)
作者(Author):
王金鑫,蒋子彦,王琪,李东,于福荣,刘海,任婧媛
WANG Jinxin,JIANG Ziyan,WANG Qi,LI Dong,YU Furong,LIU Hai,REN Jingyuan
参考文献(References):
- [1]李停.基于ARMA-GARCH模型的电价预测与研究[D].重庆:重庆师范大学,2015.
- [2]李海萍.基于ARMA-GARCH模型的短期电价预测[D].北京:华北电力大学,2009.
- [3]陆洋,忻建华,叶春.基于ARMAX模型的短期出清电价预测方法[J].华东电力,2005(10):57-59.
- [4]谢培元,曾次玲.基于神经网络的短期电价预测[J].湖南电力,2004(2):9-11.
- [5]陶雪峰,谭伦农.基于最优组合的短期电价预测研究[J].科学技术与工程,2008(14):3917-3920.
- [6]李浩威.基于云计算和BP神经网络的短期电价预测研究[D].北京:华北电力大学,2014.
- [7]王勤智.电力市场下短期电价研究[D].昆明:昆明理工大学,2013.
- [8]刘诗韵.电力市场下的短期电价预测研究[D].广州:广东工业大学,2020.
- [9]张金良.电力市场环境下的短期电价预测[M].北京:中国电力出版社,2020.
- [10]殷豪,曾云,孟安波,等.基于奇异谱分析的短期电价预测[J].电力系统保护与控制,2019,47(1):115-122.
- [11]康宏伟,李强,于硕,等.基于SA-PSO-BP神经网络算法的超短期风电出力预测[J].内蒙古电力技术,2020,38(6):64-68.
- [12]林琳,刘譞,康慧玲.基于季节分类和RBF自适应权重的并行组合电价预测[J].电子测量技术,2020,43(12):101-105.
- [13]彭谦,周晓洁,杨睿,等.泛在电力物联网环境下综合能源型售电公司参与电力市场竞争的报价策略研究[J].电网技术,2019,43(12):4337-4343.
- [14]陈静雯.基于神经网络的电力价格波动研究[D].南京:南京大学,2019.
- [15]黄元生,张利君.基于遗传算法的BP-LSSVM组合变权模型权重优化的短期电价预测研究[J].煤炭工程,2019,51(5):172-176.
- [16]姚子麟,张亮,邹斌,等.含高比例风电的电力市场电价预测[J].电力系统自动化,2020,44(12):49-55.
- [17]刘达,雷自强,孙堃.基于小波包分解和长短期记忆网络的短期电价预测[J].智慧电力,2020,48(4):77-83.
- 模拟退火算法
- BP神经网络算法
- 电力现货市场
- 电价预测
- 局部最优
simulated annealing algorithm - BP neural network algorithm
- power spot market
- electricity price forecasting
- local optimal solution