Sequential Monte Carlo hydraulic state estimation of an irrigation canal

Abstract

The estimation in real time of the hydraulic state of irrigation canals is becoming one of the major concerns of network managers. With this end in view, this Note presents a new approach based on the combination of a numerical solution of the open channel Saint-Venant PDE with a sequential Monte Carlo state-space estimation. We shall show that discharges and elevations along the canal are successfully estimated, and also that, concurrently, model parameters identification, such as the Manning–Strickler friction coefficient, can be performed.