A method of data expansion for marine propeller hydrodynamic performance based on priori knowledge and its application

被引：0
|
作者：
Xie S. [1 ,2 ]
Chen Y.-H. [1 ,2 ,3 ]
Qiang Y.-M. [1 ,2 ]
Li L. [1 ,2 ]
机构：
[1] China Ship Scientific Research Center, Wuxi
[2] Taihu Laboratory of Deepsea Technological Science, Wuxi
[3] School of Aeronautics and Astronautics, Zhejiang University, Hangzhou
来源：
Chuan Bo Li Xue/Journal of Ship Mechanics | 2024年 / 28卷 / 01期
关键词：
experience knowledge; hydrodynamic performance; machine learning; sample expansion; ship propeller; surrogate model;
D O I：
10.3969/j.issn.1007-7294.2024.01.004
中图分类号：
学科分类号：
摘要：
In recent years, more and more researchers have applied machine learning to predict the performance of ship propellers, but the prediction effectiveness of surrogate model is often affected by the quantity and quality of data used for training. At present, the quantity and quality of the ship propeller performance data are unsatisfactory, and the distribution of data corresponding parameters is relatively centralized and seriously uneven. Therefore, these facts may affect the accuracy and reliability of surrogate models. In order to solve this problem, this paper presents a sample expansion method based on empirical knowledge, and applies it to the prediction of ship propeller hydrodynamic performance. The results show that the sample expansion method can generate the data sample quickly, and improve the reliability and accuracy of the forecasting surrogate model. © 2024 China Ship Scientific Research Center. All rights reserved.
引用
收藏
页码：36 / 44
页数：8
相关论文
共 18 条
  • [1] Wang Xiaoqiang, Gong Zhengqi, Development status and prospect of underwater propeller technology, China Water Transportation, 4, pp. 74-76, (2021)
  • [2] Su Yumin, Huang Sheng, Prediction of hydrodynamic performance of ship propeller by panel method, Journal of Harbin Engineering University, 22, 2, pp. 1-5, (2001)
  • [3] Ding Wenwen, Li Tao, Application of full structure network technology in propeller hydrodynamic performance prediction, Ship Science and Technology, 41, 2A, pp. 10-12, (2019)
  • [4] Li Liang, Liu Dengcheng, Zheng Chaosheng, Zhou Bin, Automatic program development and experimental verification of propeller hydrodynamic performance prediction, Proceedings of the 30th National Hydrodynamic Symposium and the 15th National Hydrodynamic Academic Conference, (2019)
  • [5] Wang Bo, Zhang Yiliang, Xiong Ying, Optimization design of hydrodynamic performance of marine propeller, Journal of Wuhan University of Technology (Traffic Science and Engineering Edition), 39, pp. 773-777, (2015)
  • [6] Zeng Zhibo, Ding enbao, Tang Denghai, Optimization design of ship propeller based on BP artificial neural network and genetic algorithm, Journal of Ship Mechanics, 14, Z1, pp. 20-27, (2010)
  • [7] Qiu Minna, Research on small sample face recognition based on sample expansion, (2014)
  • [8] Frid-Adar M, Diamant I, Klang E, Et al., GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification, Neurocomputing, 321, pp. 321-331, (2018)
  • [9] He Jiamei, Research on image expansion model and algorithm of power equipment based on generation countermeasure network, (2020)
  • [10] Araujo A., Polynomial regression with reduced over-fitting—The PALS technique, Measurement, 124, pp. 515-521, (2018)
12