文献类型：期刊文献

英文题名：Electricity load forecasting by an improved forecast engine for building level consumers

作者：Liu, Yang[1];Wang, Wei[2];Ghadimi, Noradin[3]

第一作者：刘延光

通讯作者：Ghadimi, N[1]

机构：[1]Henan Univ Econ & Law, Cloud Comp & Big Data Inst, Zhengzhou 450046, Henan, Peoples R China;[2]Henan Ind & Trade Vocat Coll, Dept Comp Sci & Technol, Zhengzhou 450003, Henan, Peoples R China;[3]Islamic Azad Univ, Young Researchers & Elite Club, Ardabil Branch, Ardebil, Iran

第一机构：河南财经政法大学

通讯机构：[1]corresponding author), Islamic Azad Univ, Young Researchers & Elite Club, Ardabil Branch, Ardebil, Iran.

年份：2017

卷号：139

起止页码：18-30

外文期刊名：ENERGY

收录：;EI(收录号：20173104010838);Scopus(收录号：2-s2.0-85026485550);WOS:【SCI-EXPANDED(收录号:WOS:000414879500003)】；

基金：This research is funded by the science and technology key project of henan province of China (No. 162102210096, No. 152102210088, No. 142102210090).

语种：英文

外文关键词：Building electricity load; Max-relevance min-redundancy; IENN; Empirical mode decomposition

摘要：For optimal power system operation, electrical generation must follow electrical load demand. So, short term load forecast (STLF) has been proposed by researchers to tackle the mentioned problem. Not merely has it been researched extensively and intensively, but also a variety of forecasting methods has been raised. This paper outlines a new prediction model for small scale load prediction i.e., buildings or sites. The proposed model is based on improved version of empirical mode decomposition (EMD) which is called sliding window EMD (SWEMD), a new feature selection algorithm and hybrid forecast engine. The aims of proposed feature selection algorithm is to maximize the relevancy and minimize the redundancy criterion based on Pearson's correlation (MRMRPC) coefficient. Finally, an improved Elman neural network (IENN) based forecast engine proposed to predict the load signal in this procedure. All weights of this forecast engine have been optimized with an intelligent algorithm to find better prediction results. Effectiveness of the proposed model is carried out to real-world engineering test case in comparison with other prediction models. (C) 2017 Elsevier Ltd. All rights reserved.