基于改进遗传算法优化的小波神经网络短期负荷预测模型

Short-term Load Forecasting Model of Wavelet Neural Network Based on Improved Genetic Algorithm Optimization

针对传统小波神经网络在电力系统短期负荷预测中存在预测结果的精确度依赖初始网络参数的问题,提出了一种基于改进遗传算法优化的小波神经网络短期负荷预测模型。为了保证神经网络在训练过程中,各个层的权值和阈值按最优方向变化,将遗传算法引入小波神经网络,利用遗传算法寻优能力指导权值和阈值进行优化。将概率分布策略用于遗传算法的种群交叉和变异过程,解决遗传算法在中后期搜索精度差,收敛速度慢等问题。应用结果表明,与基本的小波神经网络的预测模型相比,在只考虑短期负荷历史数据的情况下,通过均方根误差计算比较,基于改进遗传算法优化的小波神经网络短期负荷预测模型具有更高的预测精度。

In order to improve the accuracy of short-term load forecasting and further improve the forecast accuracy of wavelet neural network model,aiming at the problem that wavelet neural network is easy to produce over-fitting and forecast results depend on initial network parameters,introducing genetic algorithm into wavelet neural network,using genetic algorithm to optimize wavelet neural network parameters,and establishing short-term load forecasting model based on improved genetic algorithm for wavelet neural network,at the same time,aiming at the problems of poor accuracy and slow convergence in the middle and late stages of genetic algorithm,a cross-mutation strategy based on probability distribution is proposed to improve the convergence speed of the algorithm.Finally,the accuracy and effectiveness of the short-term load forecasting model based on improved genetic algorithm for wavelet neural network are verified by simulation and analysis of real power load data.