蚁群优化算法
大洪水
遗传算法
计算机科学
数学优化
布线(电子设计自动化)
启发式
托普西斯
紧急疏散
运筹学
算法
工程类
人工智能
数学
计算机网络
哲学
海洋学
神学
地质学
作者
Kai Dong,Dongwen Yang,Jinbao Su,Wendong Zhang,Peiran Jing
标识
DOI:10.1016/j.ijdrr.2023.104219
摘要
As the major water storage infrastructure, the failure of a reservoir dam will cause significant losses to people's lives and property. A scientific and reasonable evacuation route is one of the essential measures to reduce casualties. Thus, it is of great significance to realize the dynamic planning of evacuation routes in dam-break flood scenarios to reduce the risk of dam failure. This study employed a physically based mathematical model (Breach model) and a two-dimensional hydrodynamic model to simulate dam-break flood routing. Further, a road network construction method was proposed based on the graph theory and flood routing information. The ant colony optimization algorithm (ACO) was improved by the backtracking, the improved heuristic function, and the elite ants, and the genetic algorithm (GA) was optimized according to the characteristics of flood avoidance route planning. Accordingly, an improved ant colony-genetic optimization hybrid algorithm (ACO-GA) was proposed. Compared with the basic ACO algorithm, the correct rate of simulation results of the ACO-GA hybrid algorithm was improved by 34 %, the average number of iterations was reduced by 18.3 times, and the optimization ability was markedly enhanced. Based on the constructed dam-break flood scenario and the ACO-GA hybrid algorithm, the novel dynamic planning method for evacuation routes in the dam-break flood scenario was proposed. Moreover, this method was applied to a typical water conservancy engineering project, and the planned optimal evacuation routes accurately avoided the flood influence, and the calculation results were reasonable and accurate. The real-time dynamic planning of flood avoidance routes based on flood routing information was realized. This study provides effective guarantee for the emergency transfer of downstream people, which has valuable implications for the theoretical research and engineering practice of reservoir dam safety and management.
科研通智能强力驱动
Strongly Powered by AbleSci AI