Structural strength analysis and optimization of converter lug based on Kriging model

HE Guanqiang; LIU Yongjiang; LI Hua; CHEN Liming

doi:10.13890/j.issn.1000-128X.2022.04.015

您当前的位置：

首页 >

文章列表页 >

Structural strength analysis and optimization of converter lug based on Kriging model

Railway Rolling Stock | 更新时间：2024-08-02

- Structural strength analysis and optimization of converter lug based on Kriging model
- Electric drive for locomotives Issue 4, Pages: 104-110(2022)
- 作者机构：
  
  株洲中车时代电气股份有限公司轨道交通技术中心，湖南株洲　　412001
- 作者简介：
- 基金信息：
- DOI：10.13890/j.issn.1000-128X.2022.04.015
  CLC： U264.3⁺7
- Published：10 July 2022，
  
  Received：27 February 2022，
  
  Revised：10 April 2022，
- 稿件说明：
扫描看全文
HE Guanqiang, LIU Yongjiang, LI Hua, et al. Structural strength analysis and optimization of converter lug based on Kriging model. [J]. Electric drive for locomotives (4):104-110(2022)
DOI：

HE Guanqiang, LIU Yongjiang, LI Hua, et al. Structural strength analysis and optimization of converter lug based on Kriging model. [J]. Electric drive for locomotives (4):104-110(2022) DOI： 10.13890/j.issn.1000-128X.2022.04.015.

摘要

开展车下安装的轨道交通变流器结构分析通常涉及耗时的仿真计算，以评估结构强度是否符合安全要求。对于典型的轨道交通变流器产品，进行1次静强度分析（4个工况）需耗时1 h以上，进行1次单方向的随机振动分析需要8 h以上。当采用传统的工程优化方法对变流器进行结构优化时，由于需要大量调用仿真进行评估，优化效率受到严重限制。因此采用机器学习方法，通过拉丁超立方采样试验设计确定仿真方案（采样位置），根据仿真样本构建Kriging代理模型（高斯过程），以近似反映设计变量与响应之间的函数关系。在Kriging模型提供信息的基础上，实现高效的灵敏度分析、设计空间探索和全局优化，为设计人员提供优化设计的参考依据与推荐方案，避免出现安全裕量过低或设计冗余过大等极端情况，在提高产品性能的同时缩短设计周期，提升产品的竞争力。基于Kriging代理模型，文章给出了进行分析与优化的完整技术路线，并对其中关键步骤（试验设计、代理模型、自适应采样、灵敏度分析、设计空间探索、全局优化等）的具体实现方法进行了详细说明。通过对某牵引变流器吊耳结构进行分析与优化，验证了所提技术路线的有效性。

Abstract

The structural analysis of rail transit converters installed under the vehicle often involves time-consuming computer simulations to evaluate the strength of the structure for safety requirements. For typical rail transit converters

each static strength analysis with four working conditions costs at least 1 hour

and each random vibration analysis with one axis costs more than 8 hours. As a result

for using traditional engineering optimization methods to optimize the converter structure

the optimization efficiency will be severely limited

since these methods need to invoke a large amount of simulations as function evaluations. In this paper

machine learning methods were considered to approximate the functional relationships between design variables and responses. Specifically

the Latin Hypercube Sampling method was used to generate the samples and build Kriging surrogate models (also known as Gaussian Processes). With the information provided by the Kriging models

the sensitivity analysis

design space exploration and global optimization processes can all be facilitated so that the designers can avoid low safety allowance and over-redundant designs while the design circle is also shortened. A complete technical route of Kriging-based analysis and optimization was provided in this paper

and the key steps (design of experiments

surrogate model

adaptive sampling

sensitivity analysis

design space exploration

global optimization

etc.) were described in detail. Finally

the proposed technical route has been verified the validity by the analysis and optimization of a traction converter lug structure.

关键词

轨道交通装备变流器吊耳结构优化Kriging代理模型机器学习高斯过程仿真

Keywords

rail transit equipmentconverter lugstructural optimizationKriging surrogate modelmachine learningGaussian processessimulation

references

孙振旭, 姚永芳, 郭迪龙, 等. 高速列车气动外形优化研究进展[J]. 力学学报, 2021, 53(1): 51-74.

SUN Zhenxu, YAO Yongfang, GUO Dilong, et al. Research progress in aerodynamic optimization of high-speed trains[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(1): 51-74.

于梦阁, 潘振宽, 蒋荣超, 等. 基于近似模型的高速列车头型多目标优化设计[J]. 机械工程学报, 2019, 55(24): 178-186.

YU Mengge, PAN Zhenkuan, JIANG Rongchao, et al. Multi-objective optimization design of the high-speed train head based on the approximate model[J]. Journal of Mechanical Engineering, 2019, 55(24): 178-186.

LI Rui, XU Ping, PENG Yong, et al. Multi-objective optimization of a high-speed train head based on the FFD method[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2016, 152: 41-49.

KRIGE D G. A statistical approach to some basic mine valuation problems on the Witwatersrand[J]. Journal of the Southern African Institute of Mining and Metallurgy, 1951, 52(6): 119-139.

MATHERON G. Principles of geostatistics[J]. Economic Geology and the Bulletin of the Society of Economic Geologists, 1963, 58(8): 1246-1266.

RASMUSSEN C E, WILLIAMS C K I. Gaussian processes for machine learning[M]. Cambridge, Mass: MIT Press, 2006.

SACKS J, WELCH W J, MITCHELL T J, et al. Design and analysis of computer experiments[J]. Statistical Science, 1989, 4(4): 409-423.

CURRIN C, MITCHELL T, MORRIS M, et al. Bayesian prediction of deterministic functions, with applications to the design and analysis of computer experiments[J]. Journal of the American Statistical Association, 1991, 86(416): 953-963.

SETTLES B. Active learning literature survey[R]. Madison, WI: University of Wisconsin-Madison, 2010.

CONSTANTINE P G, DIAZ P. Global sensitivity metrics from active subspaces[J]. Reliability Engineering & System Safety, 2017, 162: 1-13.

JONES D R, SCHONLAU M, WELCH W J. Efficient global optimization of expensive black-box functions[J]. Journal of Global Optimization, 1998, 13(4): 455-492.

SHAHRIARI B, SWERSKY K, WANG Ziyu, et al. Taking the human out of the loop: a review of Bayesian optimization[J]. Proceedings of the IEEE, 2016, 104(1): 148-175.

张义, 刘永江, 范斌, 等. 新一代动力集中型动车组牵引变流器设计[J]. 机车电传动, 2021(2): 66-72.

ZHANG Yi, LIU Yongjiang, FAN Bin, et al. Design of traction converter for new generation of power centralized EMU[J]. Electric Drive for Locomotives, 2021(2): 66-72.

何艳飞, 丁杰. 基于OptiStruct的辅助变流器盖板及吊耳优化[J]. 机车电传动, 2017(2): 72-75.

HE Yanfei, DING Jie. Cover and lug optimization of auxiliary converter based on OptiStruct software[J]. Electric Drive for Locomotives, 2017(2): 72-75.

张文威, 丁杰, 邓创华. 基于EN 12663—2010标准的变流器柜体结构分析[J]. 电力机车与城轨车辆, 2020, 43(6): 55-59.

ZHANG Wenwei, DING Jie, DENG Chuanghua. Structure analysis of converter cabinet based on EN 12663—2010 standard[J]. Electric Locomotives & Mass Transit Vehicles, 2020, 43(6): 55-59.

Views

下载量

CSCD

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

Structural Optimization and Comparative Analysis of B-type Metro Bogie Frame

Study on dynamic behaviors of catenary overlap section under the impact of dropper failure

Research on fault diagnosis of EMU traction inverter based on motor vibration

Research on coordinated control method of hybrid energy storage systems connected traction power supply system based on PFTD

Related Author

Qingwei SONG

Chunyou GAO

Jing ZHANG

Zhuo CHEN

Wei WEI

SONG Qingwei

GAO Chunyou

ZHANG Jing

Related Institution

CRRC Changchun Railway Vehicles Co., Ltd.

Faculty of Transportation Engineering, Kunming University of Science and Technology

China Academy of Railway Sciences Co., Ltd.

National Innovation Center of High Speed Train

School of Remote Sensing and Information Engineering, Wuhan University

⁰