Multi-Objective Parameter Estimation of Improved Muskingum Model by Wolf Pack Algorithm and Its Application in Upper Hanjiang River, China

Abstract

In order to overcome the problems in the parameter estimation of the Muskingum model, this paper introduces a new swarm intelligence optimization algorithm—Wolf Pack Algorithm (WPA). A new multi-objective function is designed by considering the weighted sum of absolute difference (SAD) and determination coefficient of the flood process. The WPA, its solving steps of calibration, and the model parameters are designed emphatically based on the basic principle of the algorithm. The performance of this algorithm is compared to the Trial Algorithm (TA) and Particle Swarm Optimization (PSO). Results of the application of these approaches with actual data from the downstream of Ankang River in Hanjiang River indicate that the WPA has a higher precision than other techniques and, thus, the WPA is an efficient alternative technique to estimate the parameters of the Muskingum model. The research results provide a new method for the parameter estimation of the Muskingum model, which is of great practical significance to improving the accuracy of river channel flood routing.

Building an Intelligent Hydroinformatics Integration Platform for Regional Flood Inundation Warning Systems

“…Improving the Muskingum Routing method using various optimization methods such as hybrid bat-swarm algorithm [9], improved bat algorithm [10], and Wolf Pack Algorithm [11], or combined with a particle filter-based assimilation model [12] for streamflow forecasts; 5.…”

“…Following this track, hydroinformatics has emerged as an essential tool by combining science, technologies and social considerations into a holistic coherent framework to timely deal with collecting, modeling, visualizing, and sharing flood-related information and to improve the applicability and accuracy of flood warnings [27][28][29][30][31][32][33][34]. ML methods are efficient tools for extracting the key information from complex highly dimensional input-output patterns and are widely used in various hydrological problems such as flood forecasts in this special issue [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18] as well as groundwater and water management issues [35][36][37][38][39][40][41][42][43][44][45][46]. Recently, technological advances in social media have improved data gathering and dissemination, especially under the development of world-wide-web technologies.…”

Developing a novel parameter-free optimization framework for flood routing

“…Among the EMOAs are the genetic algorithm (GA) by Mohan 18 , harmony search (HS) by Kim et al 17 and by Geem et al 19 , the ant colony algorithm (ACA) by Zhan and Xu 20 , the gray-encoded accelerating genetic algorithm (GEAGA) by Chen and Yang 21 , particle swarm optimization (PSO) by Chu and Chang 22 , the immune clonal selection algorithm (ICSA) by Luo and Xie 23 , the parameter setting free harmony search (PSF-HS) algorithm by Geem 8 , the imperialist competitive algorithm (ICA) by Tahershamsi and Sheikholeslami 24 , multi-objective particle swarm optimization (MOPSO) by Azadnia and Zahraie 25 , differential evolution (DE) by Xu et al 26 , a combination of the simulated annealing (SA) algorithm and hybrid harmony search algorithm (HHSA) by Karahan et al 27 , modified honey-bee mating optimization (MHBMO) algorithm by Niazkar and Afzali 28 , the backtracking search algorithm (BSA) by Yuan et al 29 , Weed Optimization Algorithm (WOA) for extended nonlinear Muskingum model by Hamedi et al 30 , PSO for a new form of Muskingum (four-parameter Muskingum model proposed by Easa 31 ) by Moghaddam et al 32 , modified PSO by Norouzi and Bazargan 33 , hybrid modified honey-bee mating (HMHBM) algorithm by Niazkar and Afzali 34 , bat algorithm (BA) by Farzin et al 35 , wolf pack algorithm (WPA) by Bai et al 36 , shark algorithm (SA) by Farahani et al 37 . These and other algorithms have been previously applied to estimate the three parameters of the nonlinear form of the Muskingum parameters ( , , and m ).…”

Developing a Novel Parameter-free Optimization Algorithm for Flood Routing

“…Among the evolutionary or metaheuristic optimization algorithms ((EMOAs) one finds the genetic algorithm (GA) by Mohan (1997), harmony search (HS) by Geem et al (2000) and by Kim et al (2001), the ant colony algorithm (ACA) by Zhan and Xu (2005), the gray-encoded accelerating genetic algorithm (GEAGA) by Chen and Yang (2007), particle swarm optimization (PSO) by Chu and Chang (2009), the immune clonal selection algorithm (ICSA) by Luo and Xie (2010), the parameter setting free harmony search (PSF-HS) algorithm by Geem (2011), the imperialist competitive algorithm (ICA) by Tahershamsi and Sheikholeslami (2011), multiobjective particle swarm optimization (MOPSO) by Azadnia and Zahraie (2011), differential evolution (DE) by Xu et al (2012), a combination of the simulated annealing (SA) algorithm and hybrid harmony search algorithm (HHSA) by Karahan et al (2013), modified honey-bee mating optimization (MHBMO) algorithm by Niazkar and Afzali (2015), the backtracking search algorithm (BSA) by Yuan et al (2016), PSO for a new form of Muskingum (four-parameter Muskingum model proposed by Easa (2014)) by Moghaddam et al (2016), hybrid modified honey-bee mating (HMHBM) algorithm by Niazkar and Afzali (2017), bat algorithm (BA) by Farzin et al (2018), wolf pack algorithm (WPA) by Bai et al (2018), shark algorithm (SA) by Farahani et al (2019), and others. These algorithms have been applied to estimate the three parameters of the nonlinear form of the Muskingum parameters ( K ,  , and m).…”