摘要
针对传统ＢＰ学习算法的缺陷，提出了基于共轭梯度优化技术的ＡＮＮ学习算法。ＡＮＮ模型中考虑了温度、天气情况的影响，可进行工作日、一般休息日和节假日的预测。计算表明，该ＡＮＮ模型和学习算法实用、有效。
A multilayer neural network with conjugate gradient learning algorithm is developed for short term loadforecasting.The effects of temperature and weather are considered in ANN model. Loads of festival days as well as weekdaysand weekends and be forecasted based on the same ANN model. The effectiveness of the model and the learning algorithm hasbeen verified by numerical tests.
出处
《电力系统自动化》
EI
CSCD
北大核心
1999年第1期34-36,共3页
Automation of Electric Power Systems
关键词
电力系统
负荷预测
神经网络
ANN算法
power systems load forecasting artificial neural network (ANN) conjugate gradient learning algorithm
