This project is an extension of the work presented in \cite{water_control}, where a MPC control was designed for a water distribution network. In this project a nonlinear optimal control system is designed for the water distribution network. The three main objectives of the project are design of nonlinear optimal control control, design of consumer demand predictor and stability analysis the local flow control of the pumps by PI controller. The water network considered in this project consist of two pumping stations, two consumers and an elevated reservoir. For optimal control, a control structure is developed with NMPC as the supervisory control and a PI controller for local flow control of the pumps. The objectives of the NMPC are defined to be minimizing the operation cost and pressure variations at the consumer end. For the prediction of consumer demand a Kalman filter based predictor is developed, which predicts the future consumer demands based on pressure measurements of the elevated reservoir. The NMPC along with the predictor is implemented and tested on a simulated plant model and a laboratory setup, and performance is compared with the results of \cite{water_control}. For the stability analysis, first Lyapunov stability of the delay-free nonlinear model is analysed and then the stability of linearized model with delay is analysed.