基于一维卷积神经网络的浮置板钢弹簧损伤检测方法
张庆铼，薛临风
（西南交通大学 牵引动力国家重点实验室，四川 成都 610031）
摘要：钢弹簧浮置板轨道可有效缓解地铁带来的环境振动问题，但目前针对钢弹簧损伤检测方法的研究尚十分匮乏。本文提出了一种基于一维卷积神经网络（1D-CNN）的钢弹簧损伤检测方法，利用轨道板垂向加速度构建数据集，通过1D-CNN对经简单预处理的原始数据进行特征提取并对损伤情形下的数据和正常情形下的数据进行分类。为评估该方法的性能，基于车辆－浮置板轨道耦合动力学仿真生成了数据集，分析了不同运行工况对网络性能的影响，结果表明该方法具有良好的数据分类准确性。
关键词：浮置板轨道；损伤检测；车辆-轨道耦合动力学；卷积神经网络
中图分类号：U213；TP183 文献标志码：A doi：10.3969/j.issn.1006-0316.2022.01.011
文章编号：1006-0316 (2022) 01-0073-08
Damage Detection Method for Steel-spring of Floating-Slab Based on One-Dimensional Convolution Neural Networks
ZHANG Qinglai，XUE Linfeng
( State Key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu 610031, China )
Abstract：Steel-spring floating-slab track is widely applied in subway transportation to effectively alleviate the environmental vibration. However, the current research on damage detection method for the steel-spring is limited. In response to the mentioned problem, a damage detection method based on one-dimensional convolution neural networks (1D-CNN) is proposed. The vertical acceleration of the slab is employed to construct the data sets, the 1D-CNN was used for extracting features from the original data which had been preprocessed simply, and the data under damage conditions and normal conditions were classified. In order to evaluate the performance of the proposed method, a data set is generated based on the coupling dynamics simulation of the vehicle and floating-slab track. The influence of different operating conditions on the network performance is studied. The result indicates that the method has good data classification accuracy.
Key words：floating-slab track；damage detection；vehicle-track coupling dynamics；convolution neural networks
———————————————
收稿日期：2021-04-19
基金项目：国家自然科学基金（51978587，11790283，51778194）；牵引动力国家重点实验室自主课题（2019TPL-T16）
作者简介：张庆铼（1995－），男，江西上饶人，硕士研究生，主要研究方向为轨道结构损伤检测，E-mail：15520452576@163.com；薛临风（1994－），男，河南焦作人，硕士研究生，主要研究方向轨道结构服役性能。
设为首页 | 加入收藏 | 免责条款
《机械》杂志版权所有 Copyright©2008-2012 Jixiezazhi.com All Rights Reserved
电话：028-85925070 传真：028-85925073 E-mail：jixie@vip.163.com
地址：四川省成都锦江工业开发区墨香路48号 邮编：610063
Powered by PageAdmin CMS