鸡舍环境监测传感器布置方法研究

Research on the arrangement method of chicken house environmental monitoring sensor

为了探讨鸡舍温度、湿度传感器的布置数量和位置选择,从而提高对鸡舍环境监测的效率,试验将90个温度、湿度传感器分三层均匀分布在鸡舍中,分析舍内温度和湿度变化规律,并运用主成分分析法将监测结果进行线性变换,根据监测点优化率与累计方差贡献率的调和评价指标确定监测点的最佳数量,计算主成分与原始数据的累计因子负荷量并以此确定最终传感器监测点位置。结果表明:八层叠层笼养鸡舍内不同监测点温度差和湿度差变异大,分别可达5~8℃和25%~35%。基于主成分分析方法剔除邻近点后,确定了6个温度监测点(25,35,52,61,79,83号)和7个湿度监测点(1,16,32,50,52,61,79号)为最优监测点,其中52,61,79号既是温度监测点又是湿度监测点,共10个监测点,在这10个监测点处安装传感器,对应的温度和湿度累计方差贡献率分别为0.94和0.89,并通过温度和湿度的变化规律验证了选择结果的合理性。说明主成分分析法获得的监测点数量和位置可有效代表蛋鸡舍温度、湿度的空间变异性,提高蛋鸡舍环境监测效率。

In order to explore the arrangement number and location selection of chicken house temperature and humidity sensors,so as to improve the efficiency of chicken house environmental monitoring,in this experiment,90 temperature and humidity sensors were evenly distributed in three layers in the chicken house to analyze the temperature and humidity changes in the house.The principal component analysis method was used to transform the monitoring results linearly.The optimal number of monitoring points was determined according to the harmonic evaluation index of the optimization rate of monitoring points and the cumulative variance contribution rate.The cumulative factor load between principal components and raw data was calculated to determine the final sensor monitoring point location.The results showed that the temperature difference and humidity difference of different monitoring points in the eight-layer caged chicken house varied greatly,reaching 5-8℃and 25%-35%,respectively.After removing the adjacent points based on the principal component analysis method,6 temperature monitoring points(No.25,35,52,61,79,83)and 7 humidity monitoring points(No.1,16,32,50,52,61,79)were identified as the optimal monitoring points.Among them,No.52,61,79 were all temperature monitoring points and humidity monitoring points,with a total of 10 monitoring points;the sensors installed at these 10 monitoring points had the corresponding cumulative variance contribution rates of temperature and humidity of 0.94 and 0.89,respectively,and the rationality of the selection results was verified by the change law of temperature and humidity.The results suggested that the number and location of monitoring points obtained by principal component analysis could effectively represent the spatial variability of temperature and humidity in chicken houses,and improve the environmental monitoring efficiency of chicken houses.