Burst Detection in Water Distribution System Using the Extended Kalman Filter

A water distribution system burst from a pipe rupture results in water loss and disruptions of customer service. Numerous methods, including Statistical Process Control, time series modeling, and pattern recognition, have been applied to detect bursts. However, system changes its boundary conditions such as the set of operating pumps and valve closures greatly complicating the detection problem. Thus, to date applications have been limited to the network supplied by gravity or under constant boundary conditions. This study seeks to overcome these limitation using the Kalman filter method to estimate the system state and detect bursts.