基于神经网络的农业预测模型的建立

Method of Modeling of AgriculturalPrediction Based on Neural Network

以我国粮食总产的预测为例，给出了基于神经网络Ｂ－Ｐ模型的农业生产预测模型的建立方法。研究表明，用该网络建立的农业生产模型经训练后，可以得到影响粮食总产的主要因子和粮食总产之间的非线性映射关系，具有较强的容错能力、联想能力和学习能力，可以在农业生产预测模型的建立中得到广泛的应用。

Take prediction of total foodstuff production as an example, a method of modeling of agricultural prediction based on BP model is introduced in this paper. The results show that after being exercised, the network can provide nonlinear mapping relation between independent variables and dependent variable of foodstuff production system. The model has higher tolerance, association and studying capacity. It can be widely applied in modeling of agricultural forecast.