基于自组织算法的改进型GAANN预测模型

在传统的基于GA算法人工神经网络的基础上作了改进,将训练集分为两部分,在前一训练集训练后获得的网络基础上使用后一训练集进行进一步的训练获得更为优化的网络结构。针对复杂系统建模输入节点难以确定的问题,提出将其与自组织理论相结合,首先使用GMDH方法获得神经网络的初始化节点,然后使用训练好的神经网络模型进行预测。最后,将由此建立的预测模型应用于国家粮食产量预测,取得了令人满意的效果。

Forecasting Model Using a Hybrid GMDH and Improved Arithmetic of Neural Network Based on Genetic Algorithm

Based on traditional artificial neural network using Genetic Algorithm(GA),this paper introduces a further improved method applying two independent training sets to train the network in order to optimize the neural network structure.Aiming at the characteristics of neural network structure,a model using a hybrid GMDH and artificial neural network is established.It can make the selection of input-lay units easily and improve the ability of rate of studying and the adaptability of neural network.Finally a case implementing this model in analyzing and predicting the grain production of China is presented,and the result indicates that the combined model is an effective way to improve forecasting accuracy.