蚁群算法在生命线网络系统抗震拓扑优化中的应用

Application of Ant Colony Algorithm to Seismic Topological Optimization of Lifeline Network Systems

近年来,遗传算法和模拟退火算法已经应用于网络系统的抗震拓扑优化,蚁群算法也已经成功应用到多个组合优化问题中。首先论述了生命线网络系统抗震拓扑优化模型,然后介绍了网络抗震可靠度分析的最小路递推分解算法。通过对优化问题解形式的分析,采用二进制编码的蚁群算法对优化模型进行了求解。最后,结合实例分析,并与遗传算法和模拟退火算法的计算结果进行了对比。结果表明,蚁群算法可以作为一种新的工具进行网络系统的优化设计。

Lifeline systems, including transportation systems, water and gas transmission networks, energy supply systems etc. , are indispensable to our daily life. They are also essential for sustaining the life of city after a catastrophic event such as a strong earthquake. Up to now, evolutionary methods such as genetic algorithms and simulated annealing algorithms have been developed for the seismic topological optimal design of lifeline systems. More recently, ant colony optimization algorithms, which are the newly heuristic algorithms inspired by the foraging behavior of ants, have been successfully applied to a number of mixed optimization problems. In this paper, a seismic topological optimization model of lifeline network systems is presented firstly. Then, the minimal path-based recursive decomposition algorithm, an efficient and precise algorithm to calculate the seismic connectivity reliability of network, is introduced. According to the analysis of network optimization problem, a binary coding-based ant colony algorithm is suggested. Finally, a simple network with 17 nodes and 27 pipelines is investigated in detail. Also the results calculated using genetic algorithms and simulated annealing algorithms are compared with the results using ant colony algorithm. The results show that ant colony algorithm is encouraging and feasible to be a novel tool for topology optimization problem of lifeline systems.