基于BP神经网络技术的我国饲料产量预测

Prediction of Feed Yield in China Based on BP Neural Network Technology

畜牧业的发展与饲料生产密不可分,精准预测饲料产量对于畜牧业和饲料业的生产规划至关重要,也是实现可持续发展的重要保障。研究利用我国1990年至2022年的饲料产量面板数据,运用MATLAB软件模拟构建了BP神经网络预测模型,旨在科学预测我国2023年至2024年的饲料产量。结果表明:BP神经网络模型的预测值与实际值的相关系数高达0.999 8,对我国2022年饲料产量进行的模型验证显示预测误差仅为1.07%,说明该模型的迭代学习效果良好。预测结果显示,我国2023年和2024年的饲料产量分别为33 041.62和29 986.32万t,这一研究为我国饲料生产的可持续发展提供重要的理论参考。

The development of animal husbandry is inseparable from feed production,and the accurate prediction of feed production is essential for production planning of animal husbandry and feed industry,which is also an important guarantee for sustainable development.the panel data of feed production was used in China from 1990 to 2022 and the BP neural network prediction model was simulated by using MATLAB software aiming to scientifically predict the feed production in China from 2023 to 2024.The results showed that the correlation coefficient between the predicted value and the actual value of the BP neural network model was as high as 0.9998,and the model verification of China's feed yield in 2022 showed that the prediction error was only 1.07%,the indicating that the iterative learning effect of the model was good.The forecast results showed that China's feed output in 2023 and 2024 would be 330.4162 million tons and 299.8632 million tons,respectively.This study was expected to provide an important theoretical reference for the sustainable development of feed production in China.