Iterative Parameter Selection Based Artificial Neural Network for Water Quality Prediction in Tank-Cultured Aquaculture System

Abstract

Water Quality plays an important role in attaining a sustainable aquaculture system, its cumulative effect can be detrimental to the aquatic organisms as well as the environment, which in turn leads to poor growth, increased diseases and production losses. The amount of dissolved oxygen alongside other parameters such as Temperature, pH, Alkalinity and Conductivity are used to estimate the water quality index in aquaculture. There exist different approaches for the estimation of the quality index of the water in the aquatic environment. One of such approaches is the use of the Artificial Neural Network (ANN) in the prediction of this Index, however, its efficacy lies in the ability to select and use optimal parameters for the network. Thus, this work proposes the development of an Iterative Parameter Selection (IPS) algorithm for the selection optimal network parameters for the ANN such as the number of neurons in the hidden neurons. The performance of the proposed algorithm on a typical BP-ANN was evaluated using the Mean Square Error (MSE), and the Nash-Sutcliffe Efficiency (NSE) metrics. Furthermore, a comparison of the proposed algorithm with two other known algorithm shows the proposed IPS has having a better performance. Thus, this demonstrates the capability of the IPS algorithm in obtaining optimal ANN parameters for effectively determining water quality index in Aquaculture system.