Evaluation of water treatment plant using Artificial Neural Network (ANN) case study of Pimpri Chinchwad Municipal Corporation (PCMC)

Securing safe and sustainable supplies of drinking water is a major challenge for the government and scientific community, especially in times of climate change and dynamic urban and economic development. A typical water supply system comprises an intake, transmission, treatment, and distribution works. Satisfactory performance of water treatment plant (WTP) and availability of residual chlorine in the water distribution network (WDN) is necessary for the planning, management, and operation of the water supply system. Particularly in India, water shortages and poor water quality continue to be a major challenge in domestic and industrial sectors. The performance of the WTP depends on the raw water quality and unit processes involved in water treatment. The raw water quality fluctuates due to daily and seasonal changes in weather conditions, variations in water dam level, demands by consumers, heavy rainfall floods, industrial effluents, and agricultural activities. In the present paper, performance evaluation of WTP in farthest zones in WDN, namely Sant Tukaram Nagar at Nigdi WTP, Pimpri Chinchwad Municipal Corporation (PCMC), Maharashtra, India, is investigated. Artificial Neural Network (ANN) was implemented to predict the performance of WTP with correlation coefficient (R) 0.986. Feed Forward Neural Network (FFNN) Water Quality Model was developed using the Levenberg–Marquardt Training Algorithm and Bayesian Regularization Training Algorithm.