An Improved Optimization Method on the Material Mixture Ratio of Ship Stern Bearing

Abstract:

An improved optimization method on the material mixture ratio of ship stern bearing is proposed. During the optimized experiment of ship stern bearing material mixture ratio, uniform design and partial least-squares regression are used to extract the regression function between the optimization object factors (including friction coefficient of the stern bearings, abrasion loss in unit time)which can evaluate the performance of ship stern bearing material mixture ratio and the other experimental factors(including the material mixture ratios of stern bearing components, stern bearing processing temperature).The total optimization object function is composed by the proper weighting to these obtained regression functions.The optimized material mixture ratio will be attained when the comprehensive evaluation index which is the value of the total optimization object function gets its optimum value. Through an analytical comparison with existing optimization method , the improved optimization method can greatly reduce testing times and the experimental cost, shorten the experiment period in the allowed precision range. Based on the experimental analysis of a technology simulation model of related ship stern bearing, it shows that the improved optimization method is practicable and effective.

• Periodical: Advanced Materials Research (Volumes 291-294)
• Pages: 273-277
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.291-294.273
• Online since: July 2011
• Authors: Wen Li, You An Xiao
• Keywords: Material Mixture Ratio, Optimization Design, Partial Least Square (PLS), Stern Bearing, Uniform Design

References

[1] Jing Liu, Shi Liu: High Technology Letters.Vol.18 (2008), pp.59-64 (In Chinese)

[2] Kaitai Fang: Uniform Design and Uniform Design Table (Science Press, Beijing 1994). (In Chinese)

[3] Huiwen Wang, Zaibin Wu, Jie Meng: Partial least-Squares Regression—Linear and Nonlinear Methods (National Defense Industry Press, Beijing 2006). (In Chinese)

[4] Huiwen Wang: Partial Least-Squares Regression Method and Applications (National Defense Industry Press, Beijing 1999). (In Chinese)

[5] Yin Li, Qiang Liu, Huiwen Wang, "The application of related partial least square methods in multiple system modeling and analysis" in Proceedings of the International Conference on Virtual Reality and its Application in Industry,Hangzhou,China,pp.790-795, 2002.

[6] Frank I E: Chemo metrics and Intelligent Laboratory Systems. Vol.27 (1995), pp.1-9

[7] Bo Cheng, XiZhi Wu: Advances in Mathematics.Vol.28 (1999), p.375

[8] Ling Yu, Tiejun Wu, "Boosting the PLS algorithm for regressive modeling" in Proceedings of the World Congress on Intelligent Control and Automation (WCICA), Taiwan, pp.4833-4836, (2006)

DOI: 10.1109/wcica.2006.1713302

[9] Runda Jia, Zhizhong Mao, "Yuqing Chang. A nonlinear robust partial least squares method with application" in 2010 Chinese Control and Decision Conference, Xuzhou, China, pp.2334-2339, (2010)

DOI: 10.1109/ccdc.2010.5498819