An inverse problem for modeling open channel flow with movable bed

Abstract

In this research, an explicit finite difference two steps scheme developed by Richtmeyer is presented for simulation of 1D steady /unsteady flow and bed morphology in alluvial channels. Some laboratory experiments were performed in a mobile-bed flume for both of steady an unsteady flow to validate the proposed model. For improving the simulation, firstly, the role of sediment transport formulas, coupled-uncoupled approaches and simplification in the mass continuity equation are investigated. Compares between experimental data and model performance highlight the advantage of coupled method over the uncoupled method, domain role of sediment transport model in the bed morphological simulation and inaccuracy of aggradation processes due to simplifying the mixture continuity equation. Secondly the importance of changing alluvial roughness is established by testing the model with calibrated and optimized friction factors. The inverse problem of estimating alluvial flow roughness is solved using a genetic algorithm (GA) optimization model coupled with alluvial flow model. The study demonstrates that the application of GA in the search for optimal values of roughness coefficients can significantly reduce computational errors and improve the computed water stage hydrographs. Role of the choice of the objective function is also mentioned.