基于气象负荷因子的Elman神经网络短期负荷预测

Application of a Weather Component Based Elman Neural Network to Short-Term Load Forecasting

针对地区电网负荷易受气候影响的特点 ,引入气象负荷因子 ,提出了一种综合考虑各项气象因素 ,采用Elman反馈神经网络的短期负荷预测模型。由于 Elman神经网络具有动态递归性能 ,可增强负荷预测模型的适应性。经上海电网实际数据的预测仿真计算 ,证明此方法与传统神经网络预测模型相比 ,既能减少输入变量个数 ,又能有效地提高预测精度。

According to the influence of weather factors on load this paper provides a shor t-term load forecasting model based on Elman neural network with a weather comp onent.Because of its inherent dynamic behavior and robustness it is proved by si mulation results that this model has a good performance in decreasing number of inputs as well as increasing forecasting accuracy.