摘要
针对当前人工神经网络学习算法存在的问题,使用变步伐最速下降法和共轭梯度法的混合算法来进行神经网络的训练,并建立了负荷预测的人工神经网络模型。介绍了基于Delphi下的短期电力负荷预测系统。该系统由负荷预测数据查询模块、预测方法模块、结果查询模块和图表输出模块四部分组成。事实说明,混合算法在全局收敛性和收敛速度上要好于传统的算法,所基于此的短期负荷预测系统能达到令人满意的精度。
In view of the existing problems of current artificial neural network learning algorithm,a hybrid algorithm,which combines the conjugate gradient algorithm with the steepest decent gradient algorithm adopting variable step-length is presented to train the neural network.A power load forecasting model based on artificial neural network is also established.A Delphi based system for short dated forecasting is developed.The system consists of one data inquiry module,one method module,one result inquiry module and one chart output module.The results of numerical simulations show that the hybrid algorithm is better than conventional algorithm as far as the global convergency and the convergent speed goes.The precision of the forecasting system is satisfactory.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第7期216-217,232,共3页
Computer Engineering and Applications
