基于遗传BP神经网络模型的土地利用变化预测模型研究

针对已有的遗传BP神经网络土地利用变化预测模型存在BP神经网络隐层节点不易确定、创建过程烦琐等问题,本文利用输入层与隐藏层神经节点数量关系原理确定隐层节点,在Sheffield工具箱环境下进行遗传算法的编程,简化遗传BP神经网络土地利用变化预测模型的创建。结果表明,利用输入层和隐含层节点数量关系创建的遗传BP神经网络土地利用变化预测模型,可以实现土地利用变化的预测,而且在效率和精度上均优于传统BP神经网络模型,且操作简便。

Prediction Model of Land Use Change Based on Genetic BP Neural Network Model

Aiming at the existing genetic BP neural network land use change prediction model,there are some problem such as the hidden layer nodes of BP neural network are not easy to be determined with a cumbersome creating process.In this paper,the hidden layer node is determined by the theory about the relationship between the input layer and the hidden layer neural node number,Programming in the environment of Sheffield toolbox genetic algorithm,to establish a simplified genetic BP neural network model for prediction of land use change.The results show that the prediction model of land use change can be realized by using the genetic BP neural network based on the relationship between the input layer and the hidden layer nodes.And it is superior to the traditional BP neural network model in efficiency and precision,and the operation is simple and convenient.