随着我国肉牛规模化设施养殖模式的逐步普及，对牛只的精细化甚至智慧化管控已提上议事日程；同时，物联网、大数据、人工智能（AI）甚至大模型的快速发展正不断渗透到各行各业，使得对包括传统养殖在内的智慧管控也成为可能。本研究以阳信亿利源5G数字化牧场为研究对象，集成应用智能电子耳标及智能项圈，以及包括温湿度、氨气、二氧化碳、风向风速、光照及空气质量（H₂S，PM2.5，PM10，TSP）等多种环境传感器。同时进行全面动态感知牛只个体的生理指标如运动量、反刍量等，并通过对上述2个重要的生理指标的分析，判断繁殖母牛的发情情况并预测最佳的配种时间，或通过监测牛只反刍量的变化，判断饲料的调整是否合适；全面感知牛舍包括温湿度及空气质量指标，为牛舍的精准通风及环境质量的精准控制提出数据支撑；并采用MY SQL数据库技术及DELPHI语言进行相关数据的分析。本研究构建了集养殖环节关键数据自动采集、数据的自动转换计算、数据的有线及无线传输、数据的远程贮存及处理控制为一体的物联网管控平台，实现了对肉牛养殖环境、健康状态、防疫及饲料调度等数字化管控平台。随着系统不断运行与迭代、养殖过程数据维度及数据量的不断积累与拓展，基础数据及派生数据的挖掘及升值空间越大，将为建立肉牛养殖的大模型奠定基础。

As large-scale beef cattle facilities become more and more common in China, sophisticated and even intelligent management of cattle has gained prominence. In the meantime, the Internet of Things, big data, artificial intelligence (AI) and even large models are developing at a rapid pace and are constantly permeating in all industries, making the intelligent management, including traditional breeding, possible. In this study, Yangxin Yi Liyuan 5G digital farm was used as the research object, which integrated application of intelligent electronic ear tags and intelligent collars, as well as a variety of environmental sensors for temperature and humidity, ammonia, carbon dioxide, wind direction and speed, light and air quality (H₂S, PM2.5, PM10, TSP). Simultaneously, a comprehensive dynamic perception of individual physiological indicators, such as the degree of exercise and rumination, was utilized to determine the estrus period of breeding cattle and forecast the ideal breeding period, and was employed to ascertain the estrus period of breeding cattle, forecast the ideal mating period, and determine whether the feed adjustment was necessary by monitoring variations in rumination time. Comprehensively monitoring the environmental indicators, such as temperature, humidity and air quality, is necessary to accurately manage and ventilate the cattle house. The pertinent data was analyzed using the MY SQL database technology and DELPHI language technology. In order to create a digital control platform for the breeding environment, health status, epidemic prevention, and feed management of beef cattle, this research built a digital control platform that integrated automatic collection of key data in breeding links, automatic conversion and calculation of data, wired and wireless transmission of data, remote storage and processing control of data, and so on. This study also shown that the more the system was used and improved, the more data was collected and extended in both size and type during the breeding process, and the more fundamental and derived data could be mined and appreciated. This paved way for the eventual development of a large-scale model for breeding beef cattle in the future.