Using genetic algorithms to calculate floodway stations with HEC-RAS
Date of Award
M.S. in Civil Engineering
Department of Civil and Environmental Engineering and Engineering Mechanics
Advisor: Donald V. Chase
The paper describes the current problem associated with HEC-RAS that could not provide an accurate 1.00 surcharge for every cross-section of a studied river in encroachment analysis. And also developed a computer model that uses an optimization technique called genetic algorithms to call HEC-RAS to do the encroachment analysis. Genetic algorithm has been applied widely in water resources field since several decades ago, such as in water distribution system design, modeling pipes, pumps, valves and tanks associated with minimum cost issue. This GA program is developed upon the classical process consisting of building the initial population, fitness test, selection, crossover and mutation. In this case, the decision variable of the problem is the percent of the overbank width and could express the encroachment station within the overbanks. And the objective function is to minimizing the difference between the actual surcharge and the target surcharge for every cross-section. An artificial channel with 12 cross-sections is tested for accuracy and responding time, and results have been compared with HEC-RAS method 4 results and manual solution results. Currently, the GA program still could not provide a result better than HEC-RAS provided, however, it is in good progress and is doing well on cutting modeling time and effort. To improve the program, some inspirations about build population, cross over, mutation, recording system, checking system, and bisection method obtained from the manual solution to pursue a more reasonable and closer result.
Floodplain management Simulation methods, Flood routing Simulation methods, Genetic algorithms, Hydraulics Computer programs, Civil Engineering, Water Resource Management, HEC-RAS, floodway station, encroachment station, floodplain management, FEMA
Copyright 2015, author
Yu, Lu, "Using genetic algorithms to calculate floodway stations with HEC-RAS" (2015). Graduate Theses and Dissertations. 1064.