基于GM(1,1)-BP神经网络的城市耕地数量预测研究

基于灰色理论和BP神经网络建立GM(1,1)-BP神经网络组合模型,把灰色模型的时序性、无序性等优点与BP神经网络自学习、自组织的特点相结合,对耕地数量进行组合预测。结果表明,组合模型对耕地数量的变化预测精度较单一的灰色预测法有提高,与实际值有较好的拟合度,且具有简单、易用的特点。

Study on the Forecasting Method for Urban Cultivated Land Quantity Based on GM (1, 1 )-BP Neural Network

Based on the grey theory and BP neural network,we build up the composition model of GM(1,1)-BP neural network.From combining the advantages of temporality and disorder of grey model with the characteristics of self-learning and self-organization of BP neural network,we can forecast the farmland quantity with the combining model.The results show that compared with the grey prediction method which is unitary to forecast the accuracy of the changes in the quantity of cultivated land,composition model has been improved.Besides it has a good degree of fitting the actual value.At the same time,it's simple and easy to use.