Machine Learning for the Nondestructive Prediction of Neutral Temperature in Continuous Welded Rails

IF 1 4区材料科学 Q3 MATERIALS SCIENCE, CHARACTERIZATION & TESTING

Research in Nondestructive Evaluation Pub Date : 2023-07-04 DOI:10.1080/09349847.2023.2237446

Matthew Belding, A. Enshaeian, Charles A Hager, P. Rizzo

{"title":"Machine Learning for the Nondestructive Prediction of Neutral Temperature in Continuous Welded Rails","authors":"Matthew Belding, A. Enshaeian, Charles A Hager, P. Rizzo","doi":"10.1080/09349847.2023.2237446","DOIUrl":null,"url":null,"abstract":"ABSTRACT This paper describes the application of machine learning (ML) in the framework of a data-driven nondestructive evaluation (NDE) method to estimate the rail neutral temperature (RNT) of continuous welded rails (CWR). The method consists of triggering vibration of the rail of interest and extracting the power spectral densities (PSDs) of the accelerations associated with the lowest modes of vibration. The PSDs then become the input of an ML algorithm trained to associate the PSD to longitudinal stress and then RNT. In the study presented in this article, the proposed NDE method was tested on a tangent track on wood cross-ties. Vibrations were induced with a hammer and detected with several wireless and wired accelerometers. The PSDs across the 0–700 Hz range were extracted from the time-series. These densities in both the lateral and vertical directions constituted part of the input of an artificial neural network trained and tested with experimental data. The predicted neutral temperatures showed very good agreement with the RNT estimated by an independent party and based on conventional strain-gage rosettes.","PeriodicalId":54493,"journal":{"name":"Research in Nondestructive Evaluation","volume":"39 1","pages":"121 - 135"},"PeriodicalIF":1.0000,"publicationDate":"2023-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Research in Nondestructive Evaluation","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1080/09349847.2023.2237446","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MATERIALS SCIENCE, CHARACTERIZATION & TESTING","Score":null,"Total":0}

引用次数: 0

Abstract

ABSTRACT This paper describes the application of machine learning (ML) in the framework of a data-driven nondestructive evaluation (NDE) method to estimate the rail neutral temperature (RNT) of continuous welded rails (CWR). The method consists of triggering vibration of the rail of interest and extracting the power spectral densities (PSDs) of the accelerations associated with the lowest modes of vibration. The PSDs then become the input of an ML algorithm trained to associate the PSD to longitudinal stress and then RNT. In the study presented in this article, the proposed NDE method was tested on a tangent track on wood cross-ties. Vibrations were induced with a hammer and detected with several wireless and wired accelerometers. The PSDs across the 0–700 Hz range were extracted from the time-series. These densities in both the lateral and vertical directions constituted part of the input of an artificial neural network trained and tested with experimental data. The predicted neutral temperatures showed very good agreement with the RNT estimated by an independent party and based on conventional strain-gage rosettes.

查看原文本刊更多论文

基于机器学习的连续焊轨中性温度无损预测

摘要:本文描述了机器学习(ML)在数据驱动无损评估(NDE)方法框架中的应用，用于估计连续焊轨(CWR)的钢轨中性温度(RNT)。该方法包括触发目标轨道的振动并提取与最低振动模态相关的加速度的功率谱密度(psd)。然后，PSD成为ML算法的输入，该算法将PSD与纵向应力和RNT相关联。在本文的研究中，所提出的无损检测方法在木材交联的切线轨道上进行了测试。振动由锤子引起，并由几个无线和有线加速度计检测。从时间序列中提取了0 ~ 700 Hz范围内的psd。横向和垂直方向上的这些密度构成了人工神经网络输入的一部分，并使用实验数据进行了训练和测试。预测的中性温度与一个独立机构基于传统应变计花环估计的RNT非常吻合。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Research in Nondestructive Evaluation 工程技术-材料科学：表征与测试

CiteScore

2.30

自引率

0.00%

发文量

审稿时长

>12 weeks

期刊介绍： Research in Nondestructive Evaluation® is the archival research journal of the American Society for Nondestructive Testing, Inc. RNDE® contains the results of original research in all areas of nondestructive evaluation (NDE). The journal covers experimental and theoretical investigations dealing with the scientific and engineering bases of NDE, its measurement and methodology, and a wide range of applications to materials and structures that relate to the entire life cycle, from manufacture to use and retirement. Illustrative topics include advances in the underlying science of acoustic, thermal, electrical, magnetic, optical and ionizing radiation techniques and their applications to NDE problems. These problems include the nondestructive characterization of a wide variety of material properties and their degradation in service, nonintrusive sensors for monitoring manufacturing and materials processes, new techniques and combinations of techniques for detecting and characterizing hidden discontinuities and distributed damage in materials, standardization concepts and quantitative approaches for advanced NDE techniques, and long-term continuous monitoring of structures and assemblies. Of particular interest is research which elucidates how to evaluate the effects of imperfect material condition, as quantified by nondestructive measurement, on the functional performance.