BOD MODELLING USING ARTIFICIAL NEURAL NETWORK

Civil Engineering

NATIONAL CONFERENCE ON TECHNOLOGICAL ADVANCEMENTS IN ENGINEERING

Water quality modelling is required for proper water quality conservation and management. Limitation of fresh water sources suggests the need for water quality protection because it influences the lives of millions people. Water quality in the superficial waters has started to degenerate as a result of wastewater being let go to the receiving surface water without any control. The need for increased accuracies in modelling water quality has motivated the researchers to develop innovative models. Artificial neural networks (ANN) are capable of identifying the complex nonlinear relationships between input and output data. Biochemical Oxygen Demand (BOD) is an important parameter for usage conditions of surface waters. The aim of this study is to predict one of the most important water quality parameters, BOD, with the help of water temperature, hardness, dissolved oxygen and electrical conductivity data. This study involves the application of ANN based on monthly BOD modelling of Chaliyar river. Chaliyar river is the fourth longest river in Kerala and its water quality conservation is vital. ANN based study using feed forward back propagation neural network is used in this study. Sensitivity analysis is carried out on ANN model developed. Nine Artificial neural network models are presented in this paper. The performance of the model increased when T, EC, H, DO were given as input and it is the best model among other models. From sensitivity analysis, it can be concluded that EC has a significant role in BOD prediction. ANNs were able to capture the hidden relationships among the input variables and output variable and gave accurate results.This study involves the application of ANN based on monthly BOD modelling of Chaliyar river. Results from ANN models are presented. The results obtained in this study suggest that the ANN technique provide accurate results.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.