Prediction of Residual Chlorine in Water Distribution System Using Artificial Neural Network (ANN)

Abstract

Protecting public health in rural water distribution networks requires reliable and safe disinfection. Although chlorine inactivation is still the most often utilized approach, there are a number of environmental and distribution constraints that make it difficult to determine the ideal chlorine dosage. An Artificial Neural Network based predictive model for determining the amount of Residual chlorine in the Rui and Shingave Water Supply Scheme in Maharashtra, India, is presented in this study. Field data, including variables like pH, temperature, and distance from the Elevated Storage Reservoir (ESR), were gathered from several nodes. The Min-Max approach was used to standardize the dataset, and the NeuroSolutions v6 software was used for processing. For the purpose of training a MLP (Multilayer Perceptron) model. The Levenberg–Marquardt approach was used to develop and train a Multilayer Perceptron (MLP) model, using 70% of the data for training and 30% for testing. Strong correlation coefficients, low MSE (Mean Squared Error), and minimal MAE (Mean Absolute Error) values across all phases showed the model's high level of accuracy. The results show that the ANN model can successfully learn the dynamics of chlorine degradation in the distribution network. With the help of programs like the Jal Jeevan Mission, this predictive method presents a viable way to optimize chlorine dosage in real time and increase intelligent water quality management. For increased operational efficiency in rural water systems, the methodology can be expanded and combined with SCADA or (IoT) Internet of Things based monitoring systems.