一种基于异常值检测和群体智能优化神经网络的电价混合预测模型  

A hybrid electricity price forecasting model based on outlier detection and swarm intelligence algorithm

作　　者：戴亮[1] 何欣[1] 崔力心[1] 倪赛赛[1] 

机构地区：[1]国网甘肃省电力公司电力科学研究院,甘肃兰州730050

出　　处：《电子测试》2015年第2X期54-56,共3页Electronic Test

摘　　要：为了提高短期电价预测精度,本文提出了一种将异常值检测、时间序列分析、神经网络以及群体智能算法相结合的混合算法。作为混合算法的具体实现,文中的异常值检测利用了残差比方法和正态分布方法,群体智能优化算法选取了粒子群(PSO)算法和布谷鸟(CS)算法。作为实例研究,本文将混合模型应用用于澳大利亚新南威尔士州短期电价预测中,结果表明,混合预测方法能在一定程度上提高模型的预测精度。To improve the forecasting accuracy of electricity price on the scale of short term,a hybrid method consisted of outlier detection,time series analysis,neural network and swarm intelligence algorithm is proposed in this paper.As realization to the hybrid method, residual ratio method and normal distribution method are employed in outlier detection,PSO and CS algorithm are chosen as the swarm intelligence algorithm in this paper.In case study,the hybrid method is applied to the forecasting of short term electricity price in New South Wales,Australia,the result shows that the hybrid method can improve the forecasting accuracy to some extent.

关 键 词：异常值检测 ARMA模型 BP神经网络 PSO算法 CS算法 

分 类 号：TM73[电气工程—电力系统及自动化] TP183[自动化与计算机技术—控制理论与控制工程]