Optimization of Fuller Experimental Formula Regional Coefficients by Linear Programming and Genetic Algorithm in unguaged watersheds with Spatial Data
Subject Areas :
Ebrahim Yousefi Mobarhan
1
(
Semnan Agriculture and Natural Resources Research Center
)
Ebrahim karimi sanghchini
2
(
Lorestan Agriculture and Natural Resources Research Center
)
behroz arasto
3
(
Semnan Agriculture and Natural Resources Research Center
)
Ali Asghar Hashemi
4
(
Semnan Agriculture and Natural Resources Research Center
)
Keywords: Genetic Algorithm, linear programming, fuller experimental formula and Regional coefficients,
Abstract :
One of the methods used for calculating flood peak discharge in non-statistical watersheds is experimental methods. One of the empirical methods used in this study is the Fuller method which has advantages over other empirical methods for different flood periods. In this research, we have compared the linear programming optimization techniques and genetic algorithm in optimizing the fuller experimental formula coefficients in Excel and MATLAB respectively for selected watersheds of the region. For this purpose, the statistics of maximum 24-hour discharge of 9 stations in West Azarbaijan province with a statistical period of 21 years were studied. Comparison of the results showed the superiority of the genetic algorithm method and then linear programming. The results also show that the use of smart search methods improves the performance of conventional methods significantly. For this purpose, the statistics of maximum 24-hour discharge of 9 stations in West Azarbaijan province with a statistical period of 21 years were studied. Comparison of the results showed the superiority of the genetic algorithm method and then linear programming. The results also show that the use of smart search methods improves the performance of conventional methods significantly. For this purpose, the statistics of maximum 24-hour discharge of 9 stations in West Azarbaijan province with a statistical period of 21 years were studied. Comparison of the results showed the superiority of the genetic algorithm method and then linear programming. The results also show that the use of smart search methods improves the performance of conventional methods significantly.
_||_
