羊群无感知体重监测系统的设计  被引量:1

Design of a non-perceptual weight monitoring system for sheep

在线阅读下载全文

作  者:陈凯东 杜永兴[1] 周李涌[1] 肖俊生[1] 李宝山[1] CHEN Kaidong;DU Yongxing;ZHOU Liyong;XIAO Junsheng;LI Baoshan(College of Information Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China)

机构地区:[1]内蒙古科技大学信息工程学院,内蒙古包头014010

出  处:《黑龙江畜牧兽医》2022年第19期54-58,63,140,共7页Heilongjiang Animal Science And veterinary Medicine

基  金:内蒙古自然科学基金项目(2019MS06021);内蒙古自治区研究生联合培养基地建设项目;内蒙古自治区研究生教育教学改革研究与实践项目(YJG20191012710);内蒙古自治区科技重大专项(2019ZD025)。

摘  要:为了解决羊只称重困难问题,在不影响羊群正常通行的前提下实现在自然状态下对羊群体重的日常监测,减小羊只的应激反应,试验设计了一种羊群无感知体重监测系统,以STM32单片机为核心控制器采集数据并传给上位机,原始数据经滑动平均滤波法、随机森林算法、平均差系数法处理后得到每只羊的体重,并对该系统进行了验证。结果表明:羊群无感知体重监测系统对羊只体重监测结果的相对误差小于1.5%,绝对误差小于1.00 kg,符合实际需求。说明羊群无感知体重监测系统可应用于畜牧生产中的羊体重监测。In order to solve the difficult problem of sheep weighing,to realize the daily monitoring of sheep weight in the natural state without affecting the normal passage of sheep,and to reduce the stress reaction of sheep,a sheep non-perceptual weight monitoring system was de⁃signed.STM32 microcontroller was used as the core controller to collect data and transmit them to the host computer.The body weight of each sheep was obtained by processing the original data with sliding average filtering,random forest algorithm and mean difference coefficient me⁃thod,and the system was verified.The results showed that the relative error and the absolute error of the sheep weight monitoring results ob⁃tained by the sheep non-perceptual weight monitoring system were less than 1.5%and less than 1.00 kg,respectively,which could meet the practical requirements.It indicated that the sheep non-perceptual weight monitoring system could be applied in the the research of animal weight monitoring system in livestock breeding production.

关 键 词:羊群 动态称重 滑动平均滤波法 随机森林算法 多元线性回归方程 平均差系数 

分 类 号:S818.9[农业科学—畜牧学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象