微型机与应用2016，Vol.35Issue(23)：56-58,3.DOI:10.19358/j.issn.1674-7720.2016.23.016
基于改进萤火虫算法的小波神经网络短期负荷预测方法
Short-term load forecasting method of optimized wavelet neural network based on modified firefly algorithm
摘要
Abstract
The traditional wavelet neural network is trained by the gradient descent algorithm , and the algorithm can easily lead to premature convergence and trap in local minimum , which affect the training accuracy of the network .In this paper , the firefly algorithm is used to train the wavelet neural network to search the optimal parameters of the network in the global .In order to improve the firefly algorithm ’ s ability of parameter optimization , the value of γis adjusted adaptively in the training process .At the same time , Gauss variation is used to improve the activity of firefly individuals in order to ensure the convergence speed and avoid falling into local minimum .The optimized wavelet neural net-work is applied to short-term load forecasting , and the simulation results show that the improved prediction model has strong nonlinear fitting a-bility and high precision .关键词
小波神经网络/萤火虫算法/负荷预测/全局寻优Key words
wavelet neural network/firefly algorithm/load forecasting/global optimization分类
计算机与自动化引用本文复制引用
刘丹,张腾飞..基于改进萤火虫算法的小波神经网络短期负荷预测方法[J].微型机与应用,2016,35(23):56-58,3.基金项目
国家自然科学基金项目（61105082）；江苏省“青蓝工程”基金（QL2016）；南京邮电大学“1311人才计划”基金（NY2013）；江苏省普通高校研究生科研创新计划项目 （）
