基于连续域蚁群算法的船舶纵向运动参数辨识  

Parameter Identification of Ship Longitudinal Motions Using Continuous Domains Ant Colony Algorithm

作　　者：戴运桃[1,2] 赵希人 刘利强 

机构地区：[1]哈尔滨工程大学理学院,哈尔滨150001 [2]哈尔滨工程大学自动化学院,哈尔滨150001

出　　处：《微计算机信息》2010年第16期23-24,90,共3页Control & Automation

摘　　要：将连续域蚁群算法应用于船舶纵向运动水动力参数辨识问题。算法首先在每一迭代步中从所有蚁群中选则一组种子蚁群,并在该组蚁群的信息素密度分布函数下进行采样,生成子代蚁群,最后进行蚁群选择,从而使各组蚁群不断向适应度值较高的搜索区域移动,最终收敛到最优解。对船舶纵向水动力参数辨识问题的求解结果表明,该算法能够快速地辨识出满足精度要求的船舶纵向运动水动力参数,验证了算法的有效性。A continuous domains Ant Colony Algorithm is applied in solving the problem of hydrodynamic parameter identification of ship Longitudinal Motions. The algorithm initializes the ant positions by chaos series. In each iteration, a seed ant colony is chosen. The sampling is under the pheromone density function of seed ant colony, and the child ant colonies are got. And the whole process makes each ant colony moves to the search space of higher fitness value, and converges to the optimization. The result of hydrodynamic parameter identification of ship shows that the algorithm can recognize the hydrodynamic parameter of ship Longitudinal Motions in a satisfied precision. And the algorithm is proved to be effective.

关 键 词：参数辨识 连续域蚁群 纵向运动 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]