船型优化问题的复杂度越来越大，导致昂贵数值计算量越来越大，进而优化成本越来越高，对优化技术提出了更高的要求：精简计算量，降低优化成本。本文针对高效全局优化算法进行研究和开发，对其初始样本点、加点法则以及收敛条件等进行了探讨，采用多种初始样本点数启动优化，采用“期望函数”作为单目标加点法则以及欧氏距离EIM函数作为多目标加点法则，使用遗传算法对加点函数进行寻优精确定位最佳新增样本点，并通过标准优化测试函数进行了验证。为船型优化问题提供了新的思路。
The ship hull optimization problem is becoming more and more complex, resulting in an increasing amount of expensive numerical calculations, which in turn makes the optimization cost higher and higher. This puts forward higher requirements for optimization technology, to streamline the calculation amount and reduce the optimization cost. This paper conducts research and development on efficient global optimization algorithms, discusses its initial sample points, sample point adding rules and convergence conditions, using a variety of initial sample points to start optimization. Expected Improvement as the single-objective sample adding rule and euclidean. The distance EIM function is used as a multi-objective sample adding rule, and a genetic algorithm is used to optimize the sample adding function to accurately locate the best new sample points, and is verified through a standard optimization test function. It provides new ideas for ship type optimization problems.
2024,46(1): 88-93 收稿日期：2023-10-30
DOI：10.3404/j.issn.1672-7649.2024.01.015
分类号：U661.31
作者简介：朱雨辰(2003-),男,主要从事算法研究
参考文献：
[1] JONES D R, SCHONLAU M, WELCH W J. Efficient global optimization of expensive black-box functions[J]. Journal of Global optimization, 1998, 13: 455-492.
[2] JEONG S, OBAYASHI S. Efficient global optimization (EGO) for multi-objective problem and data mining[C]//2005 IEEE Congress on Evolutionary Computation. IEEE, 2005, 3: 2138−2145.
[3] BARTOLI N, BOUHLEL M A, KUREK I, et al. Improvement of efficient global optimization with application to aircraft wing design[C]//17th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference. 2016: 4001.
[4] HE Y , SUN J , SONG P , et al. Variable-fidelity expected improvement based efficient global optimization of expensive problems in presence of simulation failures and its parallelization[J]. Aerospace Science and Technology, 2021: 106572.
[5] ZHANG Y, LI J, XU C, et al. Vehicle Aerodynamic Optimization: On a Combination of Adjoint Method and Efficient Global Optimization Algorithm[J]. SAE International Journal of Passenger Cars - Mechanical Systems, 2019, 12(2).
[6] ZUHAL L R, FAZA G A, PALAR P S, et al. Multi-Objective Kriging-Based Optimization for High-Fidelity Wind Turbine Design[C]//AIAA Scitech 2019 Forum. 2019.
[7] YE F, WANG H, LI G. Variable stiffness composite material design by using support vector regression assisted efficient global optimization method[J]. Structural and Multidisciplinary Optimization, 2017(1): 56.
[8] 王普毅, 白影春, 林程, 等. 基于EGO加点策略的动力电池包多目标优化[J]. 汽车工程, 2021, 43(10): 1457-1465.
[9] 毛星波, 盛冬发, 朱军. EGO算法在系杆拱桥成桥索力优化中的应用研究[J]. 河南科技, 2022(11): 041.
[10] 王红涛, 竺晓程, 杜朝辉. 改进EGO算法在跨声速翼型气动优化设计中的应用[J]. 上海交通大学学报, 2009(11): 1832-1836.
WANG Hong-tao, ZHU Xiao-cheng, DU Chao-hui. Application of the improved EGO algorithm in transonic airfoil aerodynamic optimization design[J]. Joural of Shanghai Jiaotong University, 2009(11): 1832-1836.
[11] 王若宇. 改进的高效全局优化算法及其在车身轻量化设计中的应用研究[D]. 长沙: 湖南大学, 2023.
[12] 詹大为. 并行EGO算法研究及其应用[D]. 武汉: 华中科技大学, 2018.
[13] ZHAN D, CHENG Y, LIU J. Expected improvement matrix-based infill criteria for expensive multiobjective optimization[J]. IEEE Transactions on Evolutionary Computation. 2017, 21(6): 956−975.
