本文以超大型油船主参数、局部参数、细部参数等特征参数为对象，将剩余阻力系数Cr、伴流分数w、兴波阻力系数CWTW、粘压阻力系数Cpv的CFD计算结果作为数据基础，采用支持向量机方法建立船型特征参数到船舶性能的代理模型。应用OAT方法和Sobol方法计算船型特征参数对船舶剩余阻力系数Cr、伴流分数w、兴波阻力系数CWTW、粘压阻力系数Cpv的一阶敏感性，评估超大型油船船型特征参数对各快速性能指标的影响。结果显示，支持向量机方法建立的代理模型在测试集预报结果相对误差约为1%～2%，首尾特征、进流段去流段特征、船尾横剖面特征等细节参数对超大型油船的快速性能影响较大。
In this paper, the main parameters, local parameters, detailed parameters and other characteristic parameters of the super large oil tanker are used as the object, and the CFD calculation results of the residual resistance coefficient Cr, the wake fraction w, the wave resistance coefficient CWTW, and the viscous pressure resistance coefficient Cpv are used as the data basis. The support vector machine method establishes a proxy model from ship characteristic parameters to ship performance. The OAT method and Sobol method are used to calculate the first-order sensitivity of ship type characteristic parameters to ship residual resistance coefficient Cr, wake fraction w, wave resistance coefficient CWTW, and viscous pressure resistance coefficient Cpv, to evaluate the speed performance of super large oil tanker ship type characteristic parameters The impact of indicators. The results show that the proxy model established by the support vector machine method has a relative error of about 1% to 2% in the prediction results of the test set. Fore and aft characteristics, flow-in and flow-out sections, stern cross-section characteristics and other detailed parameters have a greater impact on the fast performance of super large oil tankers .
2023,45(17): 15-19 收稿日期：2022-11-15
DOI：10.3404/j.issn.1672-7649.2023.17.003
分类号：U663.2
作者简介：严兴春(1976-)，男，高级工程师，从事船舶设计与管理工作
参考文献：
[1] 曾帅. 典型海况下的船舶阻力预报研究[D]. 大连: 大连海事大学, 2020.
[2] 黄超. 肥大型船舶阻力快速预报方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2014.
[3] 何清, 李宁, 罗文娟, 等. 大数据下的机器学习算法综述[J]. 模式识别与人工智能, 2014, 27(4): 327–336
[4] HADDARA M. R, WANG Y. Parametric identification of manoeuvering models for ships[J]. International Shipbuilding Progress, 1999, 46(445): 5–27
[5] HESS D. FALLER W. Simulation of ship maneuvers using recursive neural networks[C]. 23th Symposium on Naval Hydrodynamics, Washington D. C. (USA): National Academies Press, 2000: 223–242.
[6] CHUI F C, CHANG T L, et al. A recursive neural networks model for ship maneuverability prediction[C]//Proceeding Oceans MTS/IEEE Techno-Ocean, Kobe(Japan): EEEE, 2004: 1211–1218.
[7] BESNARD E, SCHMITZ A, HEFAZI H, SHINDE R. Constructive neural networks and their application to ship multi-disciplinary design optimization[J]. Journal of Ship Research, 2007.
[8] 张恒, 詹成胜, 刘祖源, 等. 基于船舶阻力性能的船型主尺度参数敏感度分析[J]. 船舶工程, 2015(6): 11–14
[9] 肖振业, 冯佰威, 刘祖源, 等. 基于支撑向量机的船舶阻力近似模型[J]. 计算机辅助工程, 2015, 24(4): 20–23
[10] 夏露, 杨梅花, 李朗, 等. 基于全局灵敏度分析方法的气动设计研究[J]. 西北工业大学学报, 2018, 36(1): 49–56
[11] 孙毅. 基于LH-OAT的推理公式法参数敏感性分析[J]. 水利发电, 2017, 43(9): 23–25
