Author(s): Zhao Zhang, Hao Wang, Hao Wang, Xiaohui Lei, Lingling Wang

Linked Author(s): Hao Wang, Hao Wang, Xiaohui Lei, Lingling Wang

Keywords: Hydrodynamic model; Data assimilation; Gate discharge coefficient; Ensemble Kalman filter;

Abstract: A water transfer project connected by open channels is usually a complex system, which consists of multiple hydraulic structures such as check gates, culverts, and inverted siphons. Various disturbances and performance degradation of these structures significantly increase the uncertainty of hydraulic parameters and hydrodynamic simulation. In order to improve the simulation accuracy, a 1D open channel hydrodynamic model coupled with the parameter assimilation technique was established, in which the Ensemble Kalman filter (EnKF) framework was adopted to identify and correct hydraulic parameters. Different from the real-time correction of hydraulic states, the hydrodynamic model was taken as the nonlinear observation operator. The established model was then used to identify the gate discharge coefficient in the middle route of South to North Water Transfer Project. The validated results show that the parameters with initial error can gradually converge to the true value during the assimilation process. Moreover, the hydrodynamic simulation results using the assimilated parameters are consistent with the observed values. The average errors of water level and discharge are less than 2 cm and 2%, respectively. Further, the proposed method could also be extended to the parameter identification and hydrodynamic simulation of other river network systems.

DOI: https://doi.org/10.3850/38WC092019-1399

Year: 2019