水力发电学报2011，Vol.30Issue(3)：5-9,194,6.
动态调整蚁群-BP神经网络模型在短期负荷预测中的应用
Short-term load prediction based on dynamic adjustment ant colony system and back propagation neural network hybrid algorithm
摘要
Abstract
In this paper, a dynamic adjustment ant colony system (DAACS) is proposed to improve shortterm load forecast accuracy, and a short-term load forecast model is developed by using the DAACS-BP hybrid algorithm of DAACS combined with a back propagation (BP) neural network that is trained by the DAACS algorithm. This model can automatically determine the parameters of neural network based on the data sample.A prediction of the Sichuan power grid loads was made by using the model and the history load data, with a comprehensive consideration of various load-impacting factors such as meteorology, weather, price and date types. The results show a faster convergence and a better forecast accuracy of the hybrid method than those of the traditional ant colony system algorithm-BP neural network or the BP neural network, and also a significant improvement on the generalization capacity of BP neural network. Thus, the hybrid algorithm enhances the efficiency of short-term load forecast of a power system.关键词
水电工程/短期负荷预测/DAACS-BP网络算法/动态调整蚁群算法/BP神经网络Key words
hydropower engineering/ short-term load forecast/ DAACS-BP hybrid algorithm/ dynamic adjustment ant colony system algorithm/ the back propagation neural network分类
建筑与水利引用本文复制引用
师彪,李郁侠,于新花,闫旺..动态调整蚁群-BP神经网络模型在短期负荷预测中的应用[J].水力发电学报,2011,30(3):5-9,194,6.基金项目
国家火炬计划基金(07C26213711606) （07C26213711606）
陕西省自然科学基础研究计划(SJ08E220) （SJ08E220）
山东省软科学基金(2009RKB190) （2009RKB190）
西安理工大学优秀博士学位论文和科学研究基金(106-210912,106-210917). （106-210912,106-210917）
