农产品市场监测预警深度学习智能预测方法  

Agricultural Market Monitoring and Early Warning:An Integrated Forecasting Approach Based on Deep Learning

作　　者：许世卫[1,3,4] 李乾川 栾汝朋 庄家煜[1,3,4] 刘佳佳[1,3,4] 熊露 XU Shiwei;LI Qianchuan;LUAN Rupeng;ZHUANG Jiayu;LIU Jiajia;XIONG Lu(Institute of Agricultural Information,Chinese Academy of Agricultural Sciences,Beijing 100081,China;Institute of Data Science and Agricultural Economics,Beijing Academy of Agricultural and Forestry Sciences,Beijing 100097,China;Key Laboratory of Agricultural Monitoring and Early Warning Technology,Ministry of Agriculture and Rural Affairs,Beijing 100081,China;Key Open Laboratory of Agricultural Monitoring and Early Warning Intelligent System,Chinese Academy of Agricultural Sciences,Beijing 100081,China)

机构地区：[1]中国农业科学院农业信息研究所,北京100081 [2]北京市农林科学院数据科学与农业经济研究所,北京100097 [3]农业农村部农业监测预警技术重点实验室,北京100081 [4]中国农业科学院农业监测预警智能系统重点开放实验室,北京100081

出　　处：《智慧农业(中英文)》2025年第1期57-69,共13页Smart Agriculture

基　　金：“十四五”国家重点研发计划项目(2022YFD1600603);农业农村部农业监测预警技术重点实验室开放课题基金(KLAMEWT202403)。

摘　　要：[目的/意义]农产品供给、消费和价格的变化直接影响市场监测和预警。随着中国农业生产方式和市场体系的转型,数据获取技术的进步使得农业数据呈现爆炸式增长。然而,农产品多品种的联动监测和预测仍面临数据复杂、模型狭窄、应变能力弱等挑战。因此,亟需构建适应中国农业数据特点的深度学习模型,以提升农产品市场的监测与预警能力,推动精准决策和应急响应。[方法]本研究应用深度学习方法,从中国多维农业数据资源实际出发,创新提出了一套不同监测预警对象条件下深度学习综合预测方法,构建了生成对抗与残差网络协同生产量模型(Generative Adversarial Network and Residual Network, GAN-ResNet)、变分自编码器岭回归消费预测模型(Variational Autoencoder and Ridge Regression, VAE-Ridge)、自适应变换器价格预测模型(Adaptive-Transformer)。为适应实际需求,研究在CAMES中采用“离线计算与可视化分离”策略,模型推理离线完成,平衡了计算复杂度与实时预警需求。[结果和讨论]深度学习综合预测方法在玉米单产、生猪消费量和番茄市场价格的预测上,均表现出显著的精度提升。GAN-ResNet生产量预测模型进行县级尺度玉米单产预测的平均绝对百分比误差(Mean Absolute Percentage Error, MAPE)为6.58%,运用VAE-Ridge模型分析生猪消费量的MAPE为6.28%,运用Adaptive-Transformer模型预测番茄价格的MAPE为2.25%。[结论]该研究提出的深度学习综合预测方法,具有较先进的单品种、多场景、宽条件下的农产品市场监测预警分析能力,并在处理不同区域多维数据、多品种替代、市场季节性波动等分析方面显示出优良的指标性能,可为中国农产品市场监测预警提供一套新的有效分析方法。[Significance]The fluctuations in the supply,consumption,and prices of agricultural products directly affect market monitoring and early warning systems. With the ongoing transformation of China's agricultural production methods and market system, advancements in data acquisition technologies have led to an explosive growth in agricultural data. However, the complexity of the data, the narrow applicability of existing models, and their limited adaptability still present significant challenges in monitoring and forecasting the in‐terlinked dynamics of multiple agricultural products. The efficient and accurate forecasting of agricultural market trends is critical for timely policy interventions and disaster management, particularly in a country with a rapidly changing agricultural landscape like Chi‐na. Consequently, there is a pressing need to develop deep learning models that are tailored to the unique characteristics of Chinese ag‐ricultural data. These models should enhance the monitoring and early warning capabilities of agricultural markets, thus enabling pre‐cise decision-making and effective emergency responses. [Methods] An integrated forecasting methodology was proposed based on deep learning techniques, leveraging multi-dimensional agri‐cultural data resources from China. The research introduced several models tailored to different aspects of agricultural market forecast‐ing. For production prediction, a generative adversarial network and residual network collaborative model (GAN-ResNet) was em‐ployed. For consumption forecasting, a variational autoencoder and ridge regression (VAE-Ridge) model was used, while price predic‐tion was handled by an Adaptive-Transformer model. A key feature of the study was the adoption of an "offline computing and visual‐ization separation" strategy within the Chinese agricultural monitoring and early warning system (CAMES). This strategy ensures that model training and inference are performed offline, with the results transmitted to the front-end system f

关 键 词：监测预警 深度学习 生产量预测 消费量预测 价格预测 生成对抗与残差网络协同生产量模型 变分自编码器岭回归消费预测模型 自适应变换器价格预测模型 

分 类 号：TP393.01[自动化与计算机技术—计算机应用技术]