|
|
论文:2020,Vol:38,Issue(3):677-684 |
|
|
引用本文： |
|
|
张泽斌, 张鹏飞, 李瑞珍. 健糞T]基于自组织映射的高维优化参变量相关性研究[J]. 西北工业大学学报 |
|
|
ZHANG Zebin, ZHANG Pengfei, LI Ruizhen. SOM-Based High-Dimensional Design Spaces Mapping for Multi-Objective Optimization[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 健糞T]基于自组织映射的高维优化参变量相关性研究 |
|
| 张泽斌, 张鹏飞, 李瑞珍 |
|
| 郑州大学 机械与动力工程学院, 河南 郑州 450001 |
| 摘要: |
| 针对多目标优化中计算量大、以及难以提取分析高维数据中的复杂非线性关系的问题，借助自组织映射方法，将隐藏的高维多属性数据特征展现在低维可视空间中。利用NSGA-Ⅱ得到多目标优化问题中的Pareto最优解集，并通过对数据进行聚类分析，从而得到高维最优解集内目标与参数的特征分布、映射关系等特性。以动静压阶梯腔滑动轴承为应用对象，以单位承载力下的摩擦功耗、温升和失稳转速为优化目标，考虑几何结构等约束条件，结合DoE构建相应的低成本、高精度的多目标Kriging代理模型。利用自组织映射方法提取和分析最优特征区域中各目标与参数之间的相关性特征以及映射关系。结果表明，在设计范围内目标与轴向封油边宽度、供油压力之间相关性较强，而与深腔深度、浅腔包角相关性较弱。此方法可更直观地服务于设计人员对于多目标高维优化设计结果参变量的择优。 |
| 关键词:
自组织映射
高维问题表达
克里金方法
帕累托前沿
多目标优化
滑动轴承
|
|
| SOM-Based High-Dimensional Design Spaces Mapping for Multi-Objective Optimization |
|
| ZHANG Zebin, ZHANG Pengfei, LI Ruizhen |
|
| School of Mechanical and Power Engineering, Zhengzhou University, Zhengzhou 450001, China |
| Abstract: |
| Multi-objective optimization can reveal the complex parameter-objective relationships in the high-dimensional design problems. However, the data-extraction and data-presentation of the high-dimensional complex nonlinear system suffers from the increasing dimensionality. Key features and data-distribution of high-dimensional design spaces:parameter and objective spaces could be obtained by using Self-Organizing Maps (SOM) method, which re-clusters the high-dimensional multi-attribute data existing on the Pareto front into several low-dimensional maps. Correlations among all the design variables can be drawn according the colorized topological structure of the maps. Under the constraints including geometric structure and operating parameters, a low-cost and high accurate Kriging surrogate model was established to optimize a hybrid sliding bearing based on the sequential design method. Correlations between 3 objectives:"friction-to-load" ratio, temperature rise, instability threshold speed and 4 design parameters were extracted by SOM. Optimal feature regions were captured and analyzed. Results show that, within the specific feasible design space, supply pressure, axial bearing land width have important impact on the selected objectives, whereas the other parameters such as deep pocket depth and shallow pocket angle have relatively limited impact. A series of corresponding design decisions and optimization results help to understand the mechanism of the hybrid sliding bearing system in a much more intuitive way. |
| Key words:
self-organizing maps
high-dimensional representation
kriging
pareto front
multi-objective optimization
sliding bearing
|
|
| 收稿日期: 2019-06-14
修回日期:
|
| DOI: 10.1051/jnwpu/20203830677 |
| 基金项目: 国家自然科学基金（51575498）资助 |
| 通讯作者:
Email： |
| 作者简介: 张泽斌(1982-),郑州大学讲师,主要从事基于代理模型的优化设计及滑动轴承稳健设计研究。
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] 刘俊, 宋文萍, 韩忠华, 等. 梯度增强的Kriging模型与Kriging模型在优化设计中的比较研究[J]. 西北工业大学学报,2015, 33(5):819-826 LIU Jun, SONG Wenping, HAN Zhonghua, et al. Comparative Study of GEK(Gradient-Enhanced Kriging) and Kriging When Applied to Design Optimization[J]. Journal of Northwestern Polytechnical University, 2015, 33(5):819-826(in Chinese)6(in Chinese)
[2] ZHANG Z, DEMORY B, HENNER M, et al. Space Infill Study of Kriging Meta-Model for Multi-Objective Optimization of an Engine Cooling Fan[C]//Proceedings of the ASME Turbo Expo 2014, Düsseldorf, Germany, 2014
[3] KONKA A, COIT D W, SMITH A E. Multi-Objective Optimizationusing Genetic Algorithms:a Tutorial[J]. Reliability Engnieering and System Safety,2006, 91:992-1007
[4] 韩忠华. Kriging模型及代理优化算法研究进展[J]. 航空学报,2016,37(11):3197-3225 HAN Zhonghua. Kriging Surrogate Model and Its Application to Design Optimization:a Review of Recent Progress[J]. Acta Aeronautica et Astronautica Sinica, 2016,37(11):3197-3225(in Chinese)
[5] HAN Zhonghua, XU Chenzhou, ZHANG Liang, et al.Efficient Aerodynamic Shape Optimization Using Variable-Fidelity Surrogate Models and Multilevel Computational Grids[J]. Chinese Journal of Aeronautics, 2020, 33(1):31-47
[6] COTTRELL M, OLTEANU M, ROSSI F, et al.Theoretical and Applied Aspects of the Self-Organizing Maps[J]. Advances in Intelligent Systems and Computing, 2016,428:3-26
[7] KOHONEN T. The Self-Organizing Map[J]. Neurocomputing, 1998, 21(1):1-6
[8] VESANTO J, HIMBERG J, ALHOMIENI E, et al. SOM Toolbox for Matlab5[R]. Report A57, Helsinki University of Technology, 2000
[9] 杨丽,佟操. 基于降维可视化与Kriging的齿轮振动可靠性分析[J]. 航空动力学报,2016,31(4):993-999 YANG Li, TONG Cao. Reliability Analysis of Gear Vibration Based on Dimensionality Reduction Visualization and Kriging[J]. Journal of Aerospace Power, 2016, 31(4):993-999(in Chinese)
[10] 张以文,项涛,郭星, 等. 基于SOM神经网络的服务质量预测[J]. 软件学报,2018,29(11):3388-3399 ZHANG Yiwen, XIANG Tao, GUO Xing, et al. Quality Prediction for Services Based on Som Neural Network[J]. Journal of Software, 2018,29(11):3388-3399(in Chinese)
[11] 李杨,郝志峰,谢光强,等. 质量度量指标驱动的数据聚合与多维数据可视化[J]. 智能系统学报,2013,8(4):299-304 LI Yang, HAO Zhifeng, XIE Guangqiang, et al. Quality-Metrics Driven Multi-Dimensional Data Aggregation and Visualization[J]. CAAI Trans on Intelligent Systems, 2013, 8(4):299-304(in Chinese)
[12] 郭红,岑少起,张绍林. 圆柱、圆锥动静压滑动轴承设计[M]. 郑州:郑州大学出版社, 2013:260-266 GUO Hong, CEN Shaoqi, ZHANG Shaolin. Cylindrical and Conical Hydrostatic Sliding Bearing Design[M]. Zhengzhou:Zhengzhou University Press, 2013:260-266(in Chinese)
[13] DELGADO S, GONZALO C, MARTINEZ E, et al. Improvement of Self-Organizing Maps with Growing Capability for Goodness Evaluation of Multispectral Training Patters, Geosciences and Temote Sensing Symposium[J]. IEEE Trans on Nueral Networks, 2004, 62(17):564-567
[14] DEB K, AGRAWAL S, PRATAP A, et al. A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization:NSGA-Ⅱ[C]//International Conference on Parallel Problem Solving from Nature, 2000:849-858
[15] 王丹丹. 独立供油径推浮环轴承性能优化分析与软件包开发[D]. 郑州:郑州大学,2010 WANG Dandan. Perormance Optimization Analysis of the Journal-Thrust Floating Ring Bearing with Independent Oil Supply and Software Development[D]. Zhengzhou:Zhengzhou University, 2010(in Chinese)
[16] 刘豪杰. 基于FLUENT的动静压轴承特性分析及实验研究[D]. 郑州:郑州大学, 2014 LIU Haojie. Performance Analysis and Experiment Research of Hybrid Bearing Based on FLUENT[D]. Zhengzhou:Zhengzhou University, 2014(in Chinese)
[17] 钟洪,张冠坤. 液体静压和动静压轴承设计使用手册[M]. 北京:电子工业出版社,2007:138-156 ZHONG Hong, ZHANG Guankun. Hydrostatic and Hydrostatic Bearing Design Handbook[M]. Beijing:Publishing House of Electronics Industry, 2007:138-156(in Chinese)
[18] 张泽斌,张鹏飞,郭红,等. Kriging序贯设计方法在滑动轴承优化中的应用[J]. 哈尔滨工业大学学报,2019,51(7):178-183 ZHANG Zebin, ZHANG Pengfei, GUO Hong, et al. Implementation of Kriging Model Based Sequential Design on the Optimization of Sliding Bearing[J]. Journal of Harbin Institute of Technology, 2019, 51(7):178-183(in Chinese)
[19] 刘俊. 基于代理模型的高效气动优化设计方法及应用[D]. 西安:西北工业大学,2015 LIU Jun. Efficient Surrogate-Based Optimization Method and Its Application in Aerodynamic Design[D]. Xi'an:Northwestern Polytechnical University, 2015(in Chinese)
[20] GIUNTA A, WOJTKIEWICZ S, ELDRED M. Overview of Modern Design of Experiments Methods for Computational Simulations[C]//41st Aerospace Sciences Meeting and Exhibit, 2003:649
[21] LEARY S, BHASKAR A, KEANE A. Optimal Orthogonal-Array-Based Latin Hypercubes[J]. Journal of Applied Statistics, 2003, 30(5):585-598
[22] 王刚成,马宁,顾解忡. 基于Kriging代理模型的船舶水动力性能多目标快速协同优化[J]. 上海交通大学学报,2018,52(6):666-673 WANG Gangcheng, MA Ning, GU Xiezhong. Fast Collaborative Multi-Objective Optimization for Hydrodynamic Based on Kriging Surrogate Model[J]. Journal of Shanghai Jiaotong, 2018, 52(6):666-673(in Chinese)
[23] 李权. 面向多维数据及微博社交网络的可视分析技术的研究[D]. 北京:清华大学,2012 LI Quan. Research on Visual Analytics for Multidimensional Data and Micro-Blogging Social Network[D]. Beijing:Tsinghua University, 2012(in Chinese) |
|
|
|
|
|
|
|
