摘要
电力系统短期负荷预测是电力系统运行管理和实时控制所必须的基本内容,预测结果的准确性对电力系统的安全、优质,经济运行具有重要意义。通过非参数预测法建立电力系统短期负荷预测模型,以此作为Elman神经网络训练的样本集,实现网络样本设计、结构设计与网络训练,充分发挥Elman神经网络动态特性,将改进的遗传算法和Elman神经网络相结合,通过选择,交叉、变异等遗传操作,实现了神经网络权值优化。采用基于遗传优化神经网络的电力系统短期负荷预测新算法,提高了负荷预报精度,具体算例证明了算法的可行性和有效性。
Short-term load forecasting is fundamental for running and controlling in power system and the veracity of forecasting result is crucial for power system's secure and favorable running. A new method for short-term load forecasting is presented based on neural network optimized by genetic algorithm(GA), short-term load forecasting model for power system is set up by non-parameters method as sample sets for Elman neural network(Elman NN).With GA's optimizing and Elman NN's dynamic feature, the weight optimization is realized by selection, crossing and mutation operations. The Simulation indicates the method is feasible and effective.
出处
《继电器》
CSCD
北大核心
2008年第9期39-42,47,共5页
Relay
关键词
遗传算法
神经网络
电力系统
短期负荷预测
优化
genetic algorithm
neural network
power system
short-term load forecasting
optimization