北方番茄日光温室温度传感器布设位置研究  

Study on Location of Temperature Sensors in Tomato Solar Greenhouse in North China

作　　者：于玉婷 柳平增[1,2,3] 王秀丽 张艳[1,2,3] 王珅[1,2,3] 宋成宝 马峰[4] YU Yuting;LIU Pingzeng;WANG Xiuli;ZHANG Yan;WANG Shen;SONG Chengbao;MA Feng(College of Information Science and Engineering,Shandong Agricultural University,Shandong Tai’an 271018,China;Key Laboratory of Huang-Huai-Hai Smart Agricultural Technology,Ministry of Agriculture and Rural Affairs,Shandong Tai’an 271018,China;Agricultural Big Data Research Center,Shandong Agricultural University,Shandong Tai’an 271018,China;Dezhou Agricultural and Rural Bureau,Shandong Dezhou 235035,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]农业农村部黄淮海智慧农业技术重点实验室,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]德州市陵城区农业农村局,山东德州235035

出　　处：《中国农业科技导报》2023年第3期107-118,共12页Journal of Agricultural Science and Technology

基　　金：山东省农业重大应用技术创新项目(SD2019ZZ019);山东省科技特派员项目(2020KJTPY078);山东省重大科技创新工程项目(2019JZZY010713);山东省重点研发计划项目(2019GNC106103);泰安市科技发展计划(引导计划)项目(2019GX020);葡萄智能化生产关键技术研发与应用示范项目(2022TZXD0011)。

摘　　要：为提高温室环境监测的精准性,降低环境感知成本,设计了多点环境监测系统。针对秋冬季温室温度管理难度大导致的番茄病害率增加等现象,对山东省番茄温室内的温度进行持续监测,依据实测数据选用高斯函数对温室温度的日变化曲线进行最小二乘多峰拟合,将离散的数据点转化为连续的高斯函数,通过积分比较得到番茄各生长周期的最佳监测位置。结果表明,以高斯函数为目标对温室温度进行多峰拟合分析,效果较显著,最小决定系数为0.919;番茄苗期与开花坐果期,监测点N_(15)、N_(10)与温室平均温度之间温度曲线积分面积的绝对差最小,因此该生长时期的最佳监测点为N_(15)和N_(10);番茄结果期,N_(10)和N3监测点所在位置最接近温室整体平均变化水平,为该周期的最佳监测位置。综上,在番茄不同生长周期中,温室中部监测点N_(10)更能代表温室整体水平,该方法为温室环境精准监测奠定了基础。In order to improve the accuracy of greenhouse environmental monitoring and reduce the cost of environmental perception,this study designed a multi-point environmental monitoring system.Aiming at the difficulty of temperature management the increase of tomato disease rate in autumn and winter,the temperature in tomato greenhouse in Shandong province was continuously monitored.Based on the measured data,gaussian function was used to fit the daily variation curve of greenhouse temperature by least square multi-peak method,and the discrete data points were converted into continuous Gaussian function,and the optimal monitoring position of tomato in each growth cycle was obtained by integral calculation and comparison.The results showed that multi-peak fitting analysis of greenhouse temperature with gaussian function as the target had significant effect,and the lowest goodness of fit was 0.919;the absolute difference of integral area of temperature curve between monitoring points N_(15) and N_(10) and the greenhouse average temperature was the smallest in seedling and flowering and fruit-setting periods of tomato,so the best monitoring points at this growth period were N_(15) and N_(10);the locations of N_(10) and N3 monitoring points at tomato fruit period were closest to the average change level of greenhouse as a whole,which were the best monitoring position at this period.In conclusion,in different growth period of tomato,the monitoring point N_(10) in the middle of the greenhouse could better represent the overall level of the greenhouse.This method laid a foundation for the accurate monitoring of the greenhouse environment.

关 键 词：日光温室 传感器布设 多峰拟合 高斯函数 

分 类 号：S214.3[农业科学—农业机械化工程]