EC-5和5TE土壤水分传感器的多因素性能测试与校正

摘要：含水率、电导率及表层温度等土壤墒情信息是进行农牧业精确灌溉作业的基础。无线传感器网络是目前广泛采用的监测土壤墒情的有效方法，其对传感器有着低成本、低功耗和可靠性的要求。针对网络整体需求，本文在对比不同土壤水分传感器后选定并测试了土壤水分传感器EC-5和5TE的工作性能。结果表明：EC-5和5TE的电工作特性均受温度、电导率和pH影响。5TE受电导率的影响要比受温度或pH的影响大，应用校正模型，三者相关决定系数R2依次可达0.918、0.971和0.906；EC-5具有与5TE一样的温度稳定性，但其受电导率或pH的影响效果与5TE的相反，校正模型相关决定系数R2分别为0.977、0.944和0.978。此外，虽然5TE自带的电导率测定的可靠性很高，但其温度测定漂移较大，线性校正后相关决定系数R2为0.993。总的来说，5TE和EC-5已能满足无线传感器网络监测土壤墒情信息的需求，但有必要根据本试验结果提供的对温度、电导率和pH的校正模型对其进行补偿校正。

Multi-factor evaluation and modeling correction of EC-5 and 5TE soil moisture content sensors

Soil property information is vital to plant growth and is a fundamental eco-system resource for terrestrial vegetation, especially soil moisture content (SMC), soil electrical conductivity (EC) and near-surface soil temperature. Wireless sensor network (WSN) technology enables monitoring soil water content changes with a high spatial and temporal resolution; however, WSN specifically requests sensors with features of low-cost, energy efficiency, dependability, and ease-of-use. In this paper, after evaluating different sensors under the demand of WSN application, a pair of ECH2O sensors were chosen and tested under experiments based on a standardized sensor characterization methodology using liquid medium. The results show that the electronics of the EC-5 and the 5TE are sensitive to temperature, electric conductivity and pH variations. In terms of the 5TE, conductivity correction is more necessary than temperature and pH, and the R2 of the conductivity correction equation reaches 0.918. However, though encountering the same temperature stability as the 5TE, EC-5 exhibits an opposite characteristic when it comes to conductivity and pH correction where pH compensation plays a more essential part with a calibrated R2 of 0.978. Moreover, the inherent temperature measuring ability of 5TE is dissatisfying and the calibration with a R2 of 0.993 is well generated, but the conductivity measuring function is well acceptable. Generally, it is concluded that both the 5TE sensor and the EC-5 sensor are suitable for WSN applications. Besides, this paper also suggested that temperature, electric conductivity and pH effects have to be compensated using appropriate correction functions respectively and various compensation models were provided accordingly.