基于物联网的日光温室环境监测在线校准

目的]温室环境条件特别是温度对于作物生长和发育具有十分显著的影响。日光温室调控的主要环境因子之一是温度。然而,自然环境下的光照对温度产生作用,影响空气温度的监测精度。大多温度传感要求将传感器置于避光处,然而实际应用中难以保证。方法]采用机器学习中的支持向量机算法(SVM),对日光温室内的温度智能监测算法进行了研究,根据光照情况对实时监测的温度数据进行校准。结果]通过与实验测量的数据进行对比分析,结果表明,所提出的监测方法可以较为准确地实时监测空气温度,从而无需使用隔热材料或者遮阳处理,就可以基于监测的数据更精确地对相应的环境因素进行调节。结论]基于该方法,可采用常用的工业设备实现温室大棚内实时温度数据的监测,既可以节约设备和人力成本,又可以为温室控制提供准确的数据。

Online Calibration of Environmental Monitoring in Solar Greenhouse Based on Internet of Things

Greenhouse environmental conditions, especially temperature, play a crucial role in crop growth and development. Temperature is one of the main environmental factors that can be regulated in the greenhouse. However, under natural environment, the light condition takes effect on temperature level, which further influences the monitoring precision of air temperature. Based on Support Vector Machine (SVM) algorithm in machine learning, an intelligent algorithm was proposed which calibrated the monitored temperature according to the illumination level. By comparing the calibrated air temperatures with the monitored data, the results indicated that the proposed method could monitor accurately the air temperature without using insulation materials or shading treatment, which was the basis for regulating the environmental factors. With this method, online temperature data could be obtained using commonly-used industrial device, which could save cost and provide accurate data for greenhouse control.