基于智能模糊推理的UUV艇型参数生成方法研究  

Intelligent fuzzy inference method for generating UUV hull parameters

作　　者：龚俊斌 王鹏九 汪皓 杨萌 GONG Junbin;WANG Pengjiu;WANG Hao;YANG Meng(Hanjiang National Laboratory,Wuhan 430060,China)

机构地区：[1]汉江国家实验室,湖北武汉430060

出　　处：《中国舰船研究》2024年第6期56-63,共8页Chinese Journal of Ship Research

基　　金：国防基础科研计划资助项目(JCKY2020206B037)。

摘　　要：[目的]旨在提出一种基于模糊推理的策略算法实现对无人潜航器(UUV)艇型设计参数的快速优化。[方法]使用模糊推理手段,在模糊化阶段将遗传算法生成的UUV设计参数作为初始解进行模糊化处理。之后将这些初始解作为训练样本,利用等间距模糊划分策略求得模糊规则前件参数,并计算所有UUV解的隶属度值。然后借助最小学习机(LLM)求解模糊规则的后件参数。再基于生成的前件参数和后件参数生成新的UUV解,并分别计算其航速和航程的评价隶属度值。最后,结合约束条件对新生成的UUV解进行检验,得到优化后的符合条件的UUV设计参数。[结果]实验结果显示,智能模糊推理方法能够于20 s内基于遗传算法生成的UUV初始艇型参数,快速推理出多条航速和航程的评价隶属度之和在170分以上的UUV艇型参数方案。[结论]所提方法有效提升了设计效率并平衡了航速与航程。研究成果可为智能化、快速化的UUV艇型参数的生成提供有益参考。[Objective]This study proposes a strategy algorithm based on the fuzzy inference method for the rapid optimization of unmanned underwater vehicle(UUV) hull design parameters.[Method]The initial solutions generated by the genetic algorithm are first fuzzified during the fuzzification stage,then used as training samples,and the antecedent parameters of the fuzzy rules are obtained using an equal interval fuzzy partition strategy with the membership values of all calculated UUV solutions.Next,a least learning machine(LLM) is employed to solve the consequent parameters of the fuzzy rules.Based on the generated antecedent and consequent parameters,new UUV solutions are created and the evaluation membership values for speed and range are calculated.Finally,these new UUV solutions are tested against the constraint conditions to obtain optimized and compliant UUV design parameters.[Results]The experimental results show that within 20 seconds,the intelligent fuzzy inference method can infer multiple UUV hull parameter schemes with a combined evaluation membership degree score for speed and range of over 170 points based on the initial UUV hull parameters generated by genetic algorithms.[Conclusion]The proposed method effectively enhances design efficiency and balances speed and range.The findings of this study can provide valuable references for the intelligent and rapid generation of UUV hull parameters.

关 键 词：船舶设计 人工智能 无人潜航器 多目标优化 模糊推理 

分 类 号：U662.2[交通运输工程—船舶及航道工程] U662.9[交通运输工程—船舶与海洋工程]