Monitoring system for electrical conductivity of greenhouse nutrient solutions based on WSN

The application of hydroponics and a substrate-based culture in a greenhouse has been paid much attention to by more and more people for its environmental protection and high efficiency in the production. In order to promote automation of the soilless culture in a greenhouse, it is necessary to monitor the electrical conductivity (EC) of the nutrient solution used in greenhouse. Therefore, a monitoring system for EC of the nutrient solution in a greenhouse based on WSN was developed. The system consisted of a measurement section, a transmission control section, and a terminal server. The measurement section, including a conductivity electrode (DJS-1C) and a digital temperature sensor (DS18B20), could measure the EC and temperature of nutrient solution and then send the data to the transmission control section. The transmission control section could collect and process the data from different sensor nodes, and then display the result on the LCD screen to satisfy the needs of human-device interaction. It could also transmit the data of EC and temperature to the terminal server. The core controller of the whole system was a JN5139 module, which was responsible for collecting data in the measurement section, connecting with a GPRS module in the transmission control section through serial ports, and controlling GPRS by using AT commands. In this way, the system realized the remote transmission and management of soilless culture information. The data transmission was performed based on the IEEE802.15.4 standard and TCP/IP protocol. To save the consumption of power, the system was developed in a star network topology, combining with regular sleep and switches of the power of information collecting modules. Calibration tests were done in the laboratory to verify the accuracy of the system. Several estimation models were built and verified such as Piecewise linear model, Power function model, Logarithmic model, and Polynomial model. The results showed that the Piecewise linear model was the best choice, and the R2 for fitting was greater than 0.97. In addition, experiments in greenhouse were conducted from June, 2012 to July, 2012 to monitor the EC of the nutrient solution of tomatoes. The precision reached to 0.01 mS/cm with a relative error of 2.10% and the range of the measurement was from 0.5 mS/cm up to 2.9 mS/cm. It was proved that the system could satisfy the requirements of real-time EC measurement in greenhouse.