BP神经网络预测在粮食种植面积影响因素分析中的应用——以黑龙江省为例  

Application of BP Neural Network Prediction to Analysis of Influencing Factors of Grain Planting Area——A Case Study of Heilongjiang Province

作　　者：朱律 ZHU Lü

机构地区：[1]上海市测绘院,上海200063 [2]自然资源部超大城市自然资源时空大数据分析应用重点实验室,上海200063

出　　处：《科技创新与应用》2023年第3期27-31,共5页Technology Innovation and Application

基　　金：上海市2021年度“科技创新行动计划”社会发展科技攻关项目资助(21DZ1204100)。

摘　　要：影响粮食种植面积的因素有很多,除粮食最低收购价政策外,还有诸如农业劳动力人口、粮食进出口贸易等其他因素。该文结合2005年前最低收购价政策未执行时的粮食种植面积,基于BP神经网络预测模型,得到2005年后假设未实施粮食最低收购价政策情况下的粮食种植面积,并分析最低收购价政策对城乡收入差距和进出口贸易等其他因素的影响,同时剔除这些干扰,得到只有政策影响时粮食种植面积的变化。然后建立评价模型,把BP神经网络预测的粮食种植面积变化差值和其他因素干扰量纳入执行效果评价系统。最后,对黑龙江省小麦水稻的实例进行评价分析,验证模型的合理性。There are many factors that affect the grain planting area. In addition to the minimum grain purchase price policy, there are many other factors, such as agricultural labor population, grain import and export trade and so on. This paper combines the grain planting area before 2005 when the minimum purchase price policy was not implemented, and based on the BP neural network prediction model, obtains the grain planting area after 2005 when the minimum purchase price policy was not implemented, and analyzes the impact of the minimum purchase price policy on the urban-rural income gap, import and export trade and other factors. Meanwhile, these interferences are eliminated, and the change of grain planting area under the influence of the policy is obtained. Then, the evaluation model is established, and the difference of grain planting area predicted by BP neural network and the interference of other factors are included in the implementation effect evaluation system. Finally, an example of wheat and rice in Heilongjiang Province was evaluated to verify the rationality of the model.

关 键 词：BP神经网络 粮食种植面积 粮食最低收购价 效果评价模型 影响因素 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]