Parameter Identification of Nonlinear Muskingum Model with Backtracking Search Algorithm

摘要: Nonlinear Muskingum model is a popular approach widely used for flood routing in hydraulic engineering. An improved backtracking search algorithm (BSA) proposed to estimate the parameters of nonlinear model. The orthogonal designed initialization population strategy and chaotic sequences are introduced improve exploration exploitation ability BSA. At same time, selection based individual feasibility violation developed ensure that computed outflows non-negative evolutionary process. Finally, three examples employed demonstrate performance comparison between results those Wilcoxon signed ranks test shows BSA outperforms particle swarm optimization, genetic algorithm, differential evolution other algorithms reported literature terms solution quality. Therefore, it reasonable draw conclusion satisfactory efficient choice parameter estimation