遗传算法优化的神经网络预测团雾

Agglomerate Fog Prediction Based on Neural Network Optimized by Genetic Algorithm

针对国内外研究中现有团雾预测方式中出现的不足,建立了基于遗传算法优化的神经网络预测模型,用以对高速公路团雾的发生进行预测。在利用遗传算法得到BP神经网络的初始权值和阈值基础上,通过神经网络对输入的历史团雾气象数据进行学习训练,建立团雾预测模型。经优化的神经网络模型避免了由于神经网络初始权值、阈值难以确定所造成的网络震荡问题,以及神经网络计算过程中易陷入局部解的问题。实验结果表明,优化后的团雾预测模型具有较高的预测精度,为高速公路团雾的预测提供了新的方法与思路。

In view of the shortcomings of the existing agglomerate fog prediction methods at home and abroad,establishing a neural network prediction model based on genetic algorithm optimization to predict the occurrence of expressway agglomerate fog.Based on the initial weights and thresholds of BP neural network obtained by genetic algorithm,the input historical fog meteorological data are trained by neural network,and the agglomerate fog prediction model is established.The optimized neural network model avoids the network oscillation problem caused by the difficulty in determining the initial weights and thresholds of the neural network,and the problem that the neural network is easy to fall into the local solution in the calculation process.The experimental results show that the optimized agglomerate fog prediction model has high prediction accuracy,which provides a new method and idea for the prediction of expressway agglomerate fog.