南京理工大学学报（自然科学版）2018，Vol.42Issue(3)：259-265,7.DOI:10.14177/j.cnki.32-1397n.2018.42.03.001
基于核主成分分析与改进神经网络的电力负荷中期预测模型
Middle-term power load forecasting model based on kernel principal component analysis and improved neural network
摘要
Abstract
A forecasting model is proposed by combing kernel principal component analysis( KPCA) with particle swarm optimization and back propagation neural network( PSO-BPNN) to improve the level of middle-term power load forecasting. Dimensionality reduction and reconstruction of the original input space are made with the KPCA. The data set after dimensionality reduction is input to a BPNN model optimized by PSO. The average daily peak load in each month is forecasted to revise the daily load,and the daily peak forecasting load is output in the end. This model is tested using the data provided by the European Network on Intelligent Technologies ( EUNITE ) , and the mean absolute percent error(MAPE)of this model is 1.39%.关键词
核主成分分析/粒子群优化/反向传播神经网络/电力负荷/中期预测Key words
kernel principal component analysis/particle swarm optimization/back propagation neural network/power load/middle-term forecasting分类
信息技术与安全科学引用本文复制引用
孙新程,孔建寿,刘钊..基于核主成分分析与改进神经网络的电力负荷中期预测模型[J].南京理工大学学报（自然科学版）,2018,42(3):259-265,7.基金项目
国家自然科学基金(51507086) （51507086）
江苏省自然科学基金(BK20150839) （BK20150839）
