基于神经网络和相似天数法的电价预测方法研究

Research on the method for predicting electricity prices based on neural networks and similar days method

为证明神经网络及相似天数法的模型的优越性,文章使用公开数据来训练和测试网络,分析影响电价预测的因素。文章将所提出的人工神经网络模型的预测性能与相似天数法的预测性能进行了比较,显示电力市场数据的日、周平均绝对百分比误差(Mean Absolute Percentage Error,MAPE)值较小,预测均方误差(Fieldman Mean Squared Error,FMSE)小于相应值,负荷与电价之间的相关决定系数为0.6744。仿真结果表明,基于相似天数法的人工神经网络模型能够有效、准确地预测PJM市场的位置边际价格。

This article aims to demonstrate the superiority of neural networks and similar day method models,using publicly available data to train and test the network,and analyzing the factors that affect electricity price prediction.The predictive performance of the proposed artificial neural network model was compared with that of the similar day method,and it was found that the daily and weekly mean absolute percentage error values of electricity market data were small,the fieldman mean squared error of prediction was smaller than the corresponding values,and the correlation coefficient between load and electricity price was 0.6744.The simulation results show that the artificial neural network model based on the similarity day method can effectively and accurately predict the marginal price of position in the PJM market.