舰船数据采集装置的设计需兼顾装置总体质量、装置散热性能、装置结构变形及装置结构强度等多种要求，是一个典型的多目标优化设计问题。本文采用基于精英策略实数编码、带跳变算子的非支配排序多目标遗传算法（RNSGA-II-SBJG算法），开展考虑上述要求的数据采集装置总体优化设计；考虑数据采集装置工作过程中热-结构耦合效应，建立多场耦合分析模型，基于响应面方法构建代理模型，引入灰色关联度分析方法计算Pareto解集与理想解的关联度，对优化解集进行评价排序。应用上述方法对某数据采集装置进行优化设计，数值结果表明，径向基函数拟合精度高、Pareto解集分散良好，基于灰色关联度分析的优化解评价方法能够有效、合理地选取优化设计方案。
The design of ship data acquisition unit is a typical multi-objective optimization problem since a proper trade-off is necessary among different aspects, such as the device total mass, cooling performance, device deformation and device strength. In this work, the design problem is solved by means of the RNSGA-II, a variant of the Real-coded Non-dominated Sorting Genetic Algorithm (RNSGA). Considering the thermal-structure coupling effect during the working process of the unit, a multi-field coupling analysis model is established, and an agent approximate model is constructed based on radial basis functions (RBF) method. To determine the optimal design from the Pareto solution set, the grey relational analysis approach is employed. The Pareto solution set is ranked according to the degree of correlation between among alternative, the ideal and negative ideal solution. The proposed method is applied to optimize a data acquisition unit. The numerical results indicate that the RBF has good precision, the Pareto solutions are well dispersed, and the optimal solutions can be effectively identified by the grey relational analysis.
2021,43(1): 164-169 收稿日期：2020-06-22
DOI：10.3404/j.issn.1672-7649.2021.01.031
分类号：TN802
基金项目：装备发展部装备预研项目
作者简介：杨路春(1986-),男,高级工程师,主要从事多学科优化设计技术和舰船总体性能研究
参考文献：
[1] 生建友, 王明月, 李卫忠. 电子设备机箱设计[J]. 电讯技术, 2005(2)
[2] 李玲娜, 蔺佳. 基于Ansys Icepak的密闭机箱散热仿真分析[J]. 光电技术应用, 2012, 27(6)
[3] 高志巧. 星载机箱结构设计及力学分析[J]. 郑州轻工业学院学报(自然科学版), 2013, 28(6)
[4] 公茂果, 焦李成, 杨咚咚, 等. 进化多目标优化算法研究[J]. 软件学报, 2009, 20(2)
[5] SRINIVAS N, DEB K. Muiltiobjective optimization using nondominated sorting in genetic algorithms[J]. Evolutionary Computation, 1994, 2(3): 221-248
[6] Genes, cells, and organisms: The discovery and characterization of transposable elements, by Barbara McClintock. Garland Publishing, Inc., New York, 1987, [J]. 1989, 1(2): 146.
[7] RAMTEKE M, GUPTA S K. Multiobjective optimization of an industrial nylon-6 semi batch reactor using the a-jumping gene adaptations of genetic algorithm and simulated annealing[J]. Polumer Engineering & Science, 2010, 48(11): 2198-2215
[8] NAWAZ RIPON K S, KWONG S, MAN K F. A real-coding jumping gene genetic algorithm (RJGGA) for multiobjective optimization[J]. Information Sciences, 2007, 177(2): 632-54
[9] 王冬雪. 基于Ansys的热分析与耦合过程[J]. 机械工程师, 2016(8)
[10] 张成, 王霄, 王凯, 等. 基于响应曲面和遗传算法-人工神经元网络的热塑性塑料激光透射连接强度的优化[J]. 中国激光, 2011, 38(11).
