基于遗传算法和模糊聚类算法的改进BP神经网络风速预测

The improved BP network wind speed forecasting method based on gemetic algorithm and fuzzy c-means algorithm

针对BP神经网络风速预测中存在的结构不确定以及网络过度拟合的问题,利用遗传算法的全局搜索能力和模糊聚类算法的数据筛选能力,分别对BP神经网络的结构与数据进行双重优化,提出了基于遗传算法和聚类算法的改进BP神经网络风速预测方法。仿真表明,改进风速后的预测方法大大提高了风速预测的准确性。

Aiming at the structure uncertainty and network overfitting problem of BP neural network used in wind speed forecasting, by using global search ability of genetic algorithm and data filtering capability offuzzy c-means algorithm to optimize structure and data of BP neural network, the improved wind speed forecasting method is proposed based on genetic algorithm and fuzzy c-means algorithm, the simulation results show that, the improved wind speed forecasting method improves the accuracy of wind speed forecasting enormously.