Hydroponic Nutrient Detection of Lettuce Based on ISEs Array

Abstract. Current practices for managing hydroponic are typically based on controlling the electrical conductivity (EC) and the pH of the nutrient solution, which have a problem that EC provide no information on the concentrations of individual ions. The imbalance of ionic concentrations may occur by this method. In order to maintain the concentrations of the different ions in the optimum range, this work investigates the use of ion-selective electrodes (ISEs) array for monitoring nutrient solution compositions. ISE used in this study comprised a multi-channel detection platform with an array of potentiometric sensors and complex data processing by back-propagation artificial neuronal network (BP-ANN). The experiment was carried out the simultaneous online monitoring of NO₃-N, K⁺, Ca²⁺, Cl^- and pH during the lettuce growing period of 14 days. The use of the BP-ANN resulted in a slope near unity between NO₃-N, K⁺, Ca²⁺ concentrations determined by ISEs and standard methods. With the root mean square error (RMSE) of 9.71, 6.93, 20.90 mg L^-1 for NO₃-N, K⁺, Ca²⁺, respectively, the predicted concentration acquired by BP-ANN manifested higher precision than classical Nernst model. The EC and pH were also detected to verify the feasibility of the detection platform based on ISEs array. The predicted results of each ionic concentration were consistence with the variation of the EC and pH. It appropriately indicated the change of ionic concentrations in nutrient solution. Thus, the method combined with EC, pH and the multi-channel detection platform based on ISE array was beneficial to monitor the macronutrients of hydroponic solution.