基于BP-GA混合学习算法的神经网络短期负荷预测

本文提出了修正的遗传算法和BP算法相结合的短期负荷预测方法，与传统神经网络方法相比，该方法可以加快网络学习速度和提高学习精度．我们用遗传算法来训练网络参数，直到误差趋于一稳定值，然后用优化的权值进行BP算法，实现短期负荷预测．在构建网络模型时，我们考虑了气候因素的影响，并把它作为网络的一组输入点．实验结果表明基于这一方法的负荷预测系统较高的精度和实时性．

SHORT TERM LOAD FORECASTING USING A MULTILAYER NEURAL NETWORK WITH BP-GA MIXED ALGORITHMS

In this paper, a modified method (BP-GA) for short-term load forecast is presented, which can quicken the learning speed of the network and improve the predicting precision compared with the traditional artificial neural network. We use GAs to train connection weights of multi-layer feed forward neural network (BP) until the learning error has tended to stability, here, the best initial weights have been found. Then we use BP method to finish short-term load forecast process. We also consider the influence of climate for the short-term load and make it as one of the input for the BP. Experimental results show that the short-term load forecast system based on BP-GA has high precision and high learning rate.