Real-time transient data assimilation in pipe networks using an extended Kalman filter

Abstract Hydraulic transient data assimilation in pipe networks plays a critical role in monitoring the network behaviours, thereby ensuring the safety and reliability of water supply systems. However, the existing Kalman filter (KF)-based methods integrated with traditional numerical models face a severe computational burden with a significant number of state variables caused by pipe discretization. This study presents a new approach that combines an extended KF with a recently developed efficient hydraulic transient model that requires only a coarse discretization. The new method is particularly suited when the transient fluctuation is of relatively low frequency. As the number of state variables is reduced, real-time estimation of the system’s hydraulic states can be enabled, along with an enhanced accuracy of transient predictions. The proposed method was tested in two numerical pipe networks – a seven-pipe network and a 51-pipe network, with sudden changes in demand. The results indicate that the method can provide accurate estimation of transient states in real-time and has high performance and efficiency for large pipe networks.