中国粮食多因子灰色关联神经网络预测研究

针对BP神经网络多变量输入难以确定的缺点,提出了采用灰色关联分析法确定主要影响因子输入的多因子灰色关联分析神经网络预测模型,并给出了灰色关联神经网络BP预测模型的建立方法,对我国粮食生产影响因子多因子预测实证研究结果表明,用该网络建立的模型经过训练后,可得到影响粮食产出的主要影响因子及其之间的非线性关系,网络模型新颖,具有较好的预测精度及较好的预测效果,可广泛应用于各种预测研究,有较高的推广价值。

A grey relational analysis and artificial neural networks of corn production prediction in China

Aiming at the difficulties in deciding the variables of BP artificial neural network, using method of grey relational analysis to decide the input variables, this paper puts forward a grey relational analysis BP artificial neural network model, and gives its construction method. Taking prediction of China's total corn production as an example, it introduces a method of modeling of corn production prediction based on BP model, grey relational analysis and BP artificial neural network model. The results show that after being exercised, the network can provide nonlinear mapping relation between independent variables and dependent variable of corn production in China. The model is novelty, which has higher precision and good effect. It can be widely applied in modeling of many forecasting areas, and it also has high generalization value.