无人智慧农场关键技术与构建模式——以“吨半粮”无人农场为例  

Key Technologies and Construction model for Unmanned Smart Farms:Taking the"1.5-Ton Grain per Mu"Unmanned Farm as An Example

作　　者：刘力宁 张洪奇[1,3,4] 章子文 张正辉 王甲玉 李宣宣 朱珂 柳平增 LIU lining;ZHANG Hongqi;ZHANG Ziwen;ZHANG Zhenghui;WANG Jiayu;LI Xuanxuan;ZHU Ke;LIU Pingzeng(College of Information Science and Engineering,Shandong Agricultural University,Tai'an 271018,China;College of Mechanical and Electronic Engineering,Shandong Agricultural University,Tai'an 271018,China;Agricultural Big Data Research Center,Shandong Agricultural University,Tai'an 271018,China;Key Laboratory of Huang-Huai-Hai Agriculture Technology,Ministry of Agriculture and Rural Affairs,P.R.China,Tai'an 271018,China)

机构地区：[1]山东农业大学信息科学与工程学院,山东泰安271018 [2]山东农业大学机械电子与工程学院,山东泰安271018 [3]山东农业大学农业大数据研究中心,山东泰安271018 [4]农业农村部黄淮海智慧农业技术重点实验室,山东泰安271018

出　　处：《智慧农业(中英文)》2025年第1期70-84,共15页Smart Agriculture

基　　金：中央引导地方科技发展专项资金(YDZX2022073);山东省重点研发技术(2022CXGC010609);山东省重点研发计划(2022TZXD0030)。

摘　　要：[目的/意义]无人智慧农场是智慧农业的重要实践模式。本研究以山东德州“吨半粮”无人智慧农场为实验场所,攻克大田智慧农场建设中的核心技术难题,探索其建设模式与服务机制。[方法]运用物联网技术,研发了智慧农场的立体感知网络,能够高效采集并汇聚传输环境、作物长势和设备状态等关键数据。借助数据分析挖掘技术,精准提取了小麦的物候期、麦穗特征等关键表型信息。进一步结合智能农机与智能决策技术,研发了集云管控平台、智能化设备及智能农机于一体的智能控制系统。此外,依托多源数据融合、分布式计算和地理信息系统(Geographic Information System, GIS)等技术,构建了农业生产全过程智能管控平台。[结果和讨论]“吨半粮”无人智慧农场感知系统不仅提高了数据传输质量,同时可以完成麦穗、物候期等表型特征的本地分析;智能控制系统可帮助农机提升自主作业精度和灌溉、施药效率、质量,通过农业设备的改造升级实现了农场耕作、种植、管理、收获的全链条智能化管控;大数据智慧服务平台为农户提供了气象预测、灾害预警、最佳播期等农事管理服务,极大地提高了农场管理的数字化、智能化水平。实验结果表明,自组网络数据准确率保持在85%以上,无人机施药可节药55%,灌溉模型可节水20%,“济南17”和“济麦44”分别增产10.18%和7%。[结论]研究结果可为智慧农场建设提供参考和借鉴。[Objective]As a key model of smart agriculture,the unmanned smart farm aims to develop a highly intelligent and automated system for high grain yields.This research uses the"1.5-Ton grain per Mu"farm in Dezhou city,Shandong province,as the experimental site,targeting core challenges in large-scale smart agriculture and exploring construction and service models for such farms.[Methods]The"1.5-Ton grain per Mu"unmanned smart farm comprehensively utilized information technologies such as the internet of things(IoT)and big data to achieve full-chain integration and services for information perception,transmission,mining,and application.The overall construction architecture consisted of the perception layer,transmission layer,processing layer,and application layer.This architecture enabled precise perception,secure transmission,analysis and processing,and application services for farm data.A perception system for the unmanned smart farm of wheat was developed,which included a digital perception network and crop phenotypic analysis.The former achieved precise perception,efficient transmission,and precise measurement and control of data information within the farm through perception nodes,self-organizing networks,and edge computing core processing nodes.Phenotypic analysis utilized methods such as deep learning to extract phenotypic characteristics at different growth stages,such as the phenological classification of wheat and wheat ear length.An intelligent controlled system had been developed.The system consisted of an intelligent agricultural machinery system,a field irrigation system,and an aerial pesticided application system.The intelligent agricultural machinery system was composed of three parts:the basic layer,decision-making layer,and application service layer.They were responsible for obtaining real-time status information of agricultural machinery,formulating management decisions for agricultural machinery,and executing operational commands,respectively.Additionally,appropriate agricultural machinery models

关 键 词：无人智慧农场 物联网 信息采集 智能控制 大数据分析 

分 类 号：S24[农业科学—农业电气化与自动化]