基于改进蚁群算法的船舶主尺度优化

Optimization of ship principal parameters with improved ant colony algorithm

在基本蚁群算法(ACA)基础上重构解空间,并加入混沌映射机制得到改进蚁群算法(IACA).分别用该方法和基本蚁群算法求解同一船舶主尺度(船长、船宽、吃水、型深等)优化问题,并与混沌优化的结果进行对比,结果表明改进蚁群算法搜索效率更高,全局优化稳定性更强.在此基础上,就改进蚁群算法中的参数对主尺度优化的影响进行了分析,发现全局优化结果与参数的选取有直接关系,只有合理设计参数才能得到船舶主尺度优化的全局最优点.

Based on the basic algorithms （IACA）. Using ant colony algorithms（ACA） the paper proposes an improved ant colony IACA and ACA to solve the preliminary ship design problem （length, bredth, draft, depth, etc） the result shows that improved ant colony algorithms is more efficient and global optimality. Finally the paper analyzes how much some coefficients would impact optimization. It shows that the global optimality is based on choosing the reasonable parameters.