Modeling pH and Electrical Conductivity in Hydroponics using Artificial Neural Networks

Abstract

A model for predicting pH and Electrical Conductivity (EC) responses of a deep trough hydroponic system is developed. Artificial Neural Networks are used as the method of modeling. The Feedforward Neural Network Model has 9 inputs (pH, EC, nutrient solution temperature, air temperature, relative humidity, light intensity, plant age, amount of added acid and amount of added base) and two outputs (pH and EC of the next time step). The most suitable and accurate combination of network architectures and backpropagation training algorithms was the one-hidden-layer with 9 hidden nodes architecture trained with the quasi-Newton backpropagation algorithm. During the testing of the model using new input data, one step ahead predictions of pH were within 0.01 and EC within 5 microS.cm-1.