Optimization of the Nonlinear Muskingum Model Parameters for the River Routing, Tigris River a Case Study

Abstract

Flood forecasting and management are one of the most important strategies necessary for water resource and decision planners in combating flood problems. The Muskingum model is one of the most popular and widely used applications for the purpose of predicting flood routing. The particle swarm optimization (PSO) methodology was used to estimate the coefficients of the nonlinear Muskingum model in this study, comparing the results with the methods of genetic algorithm (GA), harmony search (HS), least-squares method (LSM), and Hook-Jeeves (HJ). The average monthly inflow for the Tigris River upstream at the Al-Mosul dam was selected as a case study for estimating the Muskingum model's parameters. The analytical and statistical results showed that the PSO method is the best application and corresponds to the results of the Muskingum model, followed by the genetic algorithm method, according to the following general descending sequence: PSO, GA, LSM, HJ, HS. The PSO method is characterized by its accurate results and does not require many assumptions and conditions for its application, which facilitates its use a lot in the subject of hydrology. Therefore, it is better to recommend further research in the use of this method in the implementation of future studies and applications.

Generation of Flows Applying a Simple Method of Flood Routing to Monthly Level in La Leche Basin, Peru

“…Consequently, these results of the statistical indicators point to the high accuracy that exists between the flows of the historical record of each station with respect to the generation of flows through the application of flood routing. These ranges are analogous with the results reported by Arriola et al (2021) and Kadim et al (2021).…”

The use of genetic expression programming to optimize the parameters of the Muskingum method comparison with numerical methods, Euphrates river a case study

“…The chosen factors, such as the population size, the number of generations, the types of mutations, selection, crossover, objective magnitude, and the search space constraints, determine how well a GA functions. It is crucial to define the upper and lower boundaries a genetic algorithm searches within [23][24][25][26].…”