Predicting Aquaculture Water Quality Using Machine Learning Approaches

Good water quality is important for normal production processes in industrial aquaculture. However, situ or real-time monitoring generally not available many aquacultural systems due to relatively high costs. Therefore, it necessary predict parameters aquaculture obtain useful information managing activities. This study used back propagation neural network (BPNN), radial basis function (RBFNN), support vector machine (SVM), and least squares (LSSVM) simulate including dissolved oxygen (DO), pH, ammonium-nitrogen (NH3-N), nitrate nitrogen (NO3-N), nitrite-nitrogen (NO2-N). Published data were compare the prediction accuracy of different methods. The correlation coefficients BPNN, RBFNN, SVM, LSSVM predicting DO 0.60, 0.99, respectively. pH 0.56, 0.84, 0.57. NH3-N 0.28, 0.88, 0.25, NO3-N 0.96, 0.87, predicted NO2-N with 0.08, 0.75, SVM obtained most accurate stable results, was groundwater as source water. results showed that achieved best effect 99% both published measured from a typical system. model recommended simulating systems.