Mostrar el registro sencillo del documento
Caracterización de la condición de los rieles de tren usando la función de respuesta de frecuencia y redes profundas
dc.rights.license | Atribución-NoComercial-SinDerivadas 4.0 Internacional |
dc.contributor.advisor | Restrepo Martinez, Alejandro |
dc.contributor.author | Navas Orduz, Jose Miguel |
dc.date.accessioned | 2024-05-21T14:13:35Z |
dc.date.available | 2024-05-21T14:13:35Z |
dc.date.issued | 2024-05-19 |
dc.identifier.uri | https://repositorio.unal.edu.co/handle/unal/86128 |
dc.description | Ilustraciones, gráficos |
dc.description.abstract | Debido a la crítica importancia del riel en el mantenimiento del sistema ferroviario, se hace imprescindible desarrollar un proceso que permita evaluar el estado del riel, cuantificar la severidad de sus posibles defectos y cambios en configuraciones estructurales. Este análisis es esencial para tomar medidas apropiadas, asegurando la confiabilidad y mantenibilidad del sistema en su totalidad. En este contexto, este proceso de investigación se enfoca en la implementación de una metodología que caracteriza los rieles de tren a través de su comportamiento dinámico mediante el análisis modal. El cual, aborda el análisis frecuencial a través de tres enfoques: teórico, numérico y experimental. El proceso experimental de análisis modal se realiza mediante la técnica de ensayo de martillo. Para ello, se realiza un estudio de determinación del comportamiento frecuencial del riel con respecto a cambios en la distancia entre fijaciones y a la variación de condiciones como generación de defectos. Dicho proceso implica la captura de señales mediante un acelerómetro uniaxial para la respuesta del riel y un martillo instrumentado para el impacto. Se recopilaron 45 señales para cada distancia y condición, y posteriormente se realizaron transformaciones a través de la Función de Respuesta de Frecuencia (FRF), la Transformada de Fourier (FFT) y la Transformada Continua de Wavelet (CWT). Para la interpretación y clasificación de los datos, se emplearon métodos estadísticos, como el método Z, y técnicas de aprendizaje de máquina mediante redes convolucionales profundas (CNN). Estas fueron evaluadas utilizando criterios y métricas como la exactitud, la matriz de confusión y la curva ROC. Todo esto proporcionando una metodología funcional que permite la caracterización del comportamiento frecuencial del riel de tren, considerando modificaciones tanto en el tipo de ensamble como en las variaciones de propiedades físicas. (Tomado de la fuente) |
dc.description.abstract | Due to the critical importance of railway tracks in the maintenance of the railway system, it is imperative to develop a process that allows for the assessment of the track's condition, quantification of the severity of potential defects, and changes in structural configurations. This analysis is essential for taking appropriate measures, ensuring the overall reliability and maintainability of the system. In this context, this research process focuses on implementing a methodology that characterizes train tracks through their dynamic behavior using modal analysis. This methodology addresses frequency analysis through three approaches: theoretical, numerical, and experimental. The validation of this behavior is conducted through an experimental process using the hammer test technique. The study is carried out regarding the modification in the distance between fixations and the variation of conditions to determine the corresponding changes in frequency behavior. This process involves capturing signals using a uniaxial accelerometer for track response and an instrumented hammer for impact. 45 signals were collected for each distance and condition, subsequently transformed through Frequency Response Function (FRF), Fourier Transform (FFT), and Continuous Wavelet Transform (CWT). For data interpretation and classification, statistical methods such as the Z method and machine learning techniques employing deep convolutional neural networks (CNN) were utilized. These methods were evaluated using criteria and metrics such as accuracy, confusion matrix, and ROC curve. All of this contributes to a functional methodology enabling the characterization of the frequency behavior of train tracks, considering modifications in both assembly types and variations in physical properties. |
dc.format.extent | 127 páginas |
dc.format.mimetype | application/pdf |
dc.language.iso | spa |
dc.publisher | Universidad Nacional de Colombia |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/ |
dc.subject.ddc | 380 - Comercio , comunicaciones, transporte::385 - Transporte ferroviario |
dc.subject.ddc | 620 - Ingeniería y operaciones afines::625 - Ingeniería de ferrocarriles y de carretera |
dc.subject.ddc | 620 - Ingeniería y operaciones afines::621 - Física aplicada |
dc.title | Caracterización de la condición de los rieles de tren usando la función de respuesta de frecuencia y redes profundas |
dc.type | Trabajo de grado - Maestría |
dc.type.driver | info:eu-repo/semantics/masterThesis |
dc.type.version | info:eu-repo/semantics/acceptedVersion |
dc.publisher.program | Medellín - Minas - Maestría en Ingeniería Mecánica |
dc.contributor.researchgroup | Grupo de Promoción E Investigación en Mecánica Aplicada Gpima |
dc.description.degreelevel | Maestría |
dc.description.degreename | Magíster en Ingeniería Mecánica |
dc.description.researcharea | Investigación en Ingeniería Mecánica |
dc.identifier.instname | Universidad Nacional de Colombia |
dc.identifier.repo | Repositorio Institucional Universidad Nacional de Colombia |
dc.identifier.repourl | https://repositorio.unal.edu.co/ |
dc.publisher.faculty | Facultad de Minas |
dc.publisher.place | Medellín, Colombia |
dc.publisher.branch | Universidad Nacional de Colombia - Sede Medellín |
dc.relation.indexed | LaReferencia |
dc.relation.references | M. Oregui, Z. Li , R. Dollevoet. Identification of characteristic frequencies of damaged railway tracks using field hammer test measurements. Delft, Netherlands. (2014). |
dc.relation.references | G. Idárraga, J. Sánchez, J. Santa, A. Toro. Identificación de mecanismos de desgaste en rieles de vía comercial del metro de Medellín. Medellín. (2014). |
dc.relation.references | Stuart L Grassie. Studs and squats: The evolving story. Cambridge, United Kingdom. (2016). |
dc.relation.references | M. Oregui, M. Molodova, A. Nuñez, R. Dollevoet, Z. Li. Experimental investigation into the condition of Insulated rail joints by impact excitation. Delft, Netherlands. (2015). |
dc.relation.references | Luis A. Bedian. Determinación experimental de los parámetros modales (frecuencias naturales) de una viga en condición empotrada – libre. Veracruz, México. (2010). |
dc.relation.references | A.P De Man. Dynatrack: A survey of dynamic railway track properties and their quality. Delft, Netherlands. (2002). |
dc.relation.references | Byoung‑Gyu Song, Namcheol Kang. Application of deep neural networks for the parameter identifications of lumped and distributed parameter models under severe noises and various initial values. Daegu, Republic of Korea. (2023). |
dc.relation.references | Eun-Taik Lee, Yu-Sik Hong and Hee-Chang Eun. Prediction of the physical properties of a structural member by the impact hammer test. Seoul, Korea. (2022). |
dc.relation.references | Guillermo Montiel-Varela, Alan Domínguez-Vazquez, Ezequiel Gallardo-Hernández, Luigi Bregant and Rafael García-Illescas. Experimental and numerical study for detection of rail defect. Ciudad de México, México. (2017). |
dc.relation.references | Hongyu Tao and Pan Zhang. Characterization and mitigation of wheel-rail impact at a singular rail defect. Delft, Netherlands. (2022). |
dc.relation.references | Osama Brinji, W. Kong, G. Tew. Investigating the dynamic response of the ballast. Melbourne, Australia. (2016). |
dc.relation.references | Anil Kumar, Yuqing Zhou, C.P. Gandhi, Rajesh Kumar, Jiawei Xiang. Bearing defect size assessment using wavelet transform based Deep Convolutional Neural Network (DCNN). Alexandria, Egypt. (2020). |
dc.relation.references | Yongzhi Qu, Gregory W. Vogl, Zechao Wang. A deep neural network model for learning runtime frequency response function using sensor measurements. Minnesota, USA. (2020). |
dc.relation.references | Athanasios Synodinos. Identification of railway track components and defects by analysis of wheel-rail interaction noise. Southampton, UK. (2016). |
dc.relation.references | A. Paral, D. Singha Roy, A. Samanta. A deep learning-based approach for condition assessment of semi-rigid joint of steel frame. Durgapur, India. (2020). |
dc.relation.references | J. Sresakoolchai, S. Kaewunruen. Detection and severity evaluation of combined rail defects using deep learning. Birmingham, United Kingdom. (2021). |
dc.relation.references | R. Janeluiktis. Continuous wavelet transform-based method for the enhancing estimation of wind turbine blade natural frequencies and damping for machine learning porpoises. Denmark. (2020). |
dc.relation.references | X. Du, X. Jin, G. Zhao, Z. Wen, W. Li. Rail corrugation of high-speed railway induced by rail grinding. China. (2021). |
dc.relation.references | A. Presas, D. Valentin, E. Egusquiza, C. Valero, M. Egusquiza, M. Bossio. Accurate determination of the frequency response function of submerged and confined structures by using PZT-patches. Barcelona. (2017). |
dc.relation.references | S. Rani. An experimental investigation of cantilever beam using impulse modal analysis technique. India. (2018). |
dc.relation.references | M. Loidolt, S. Marsching. Evaluating short-wave effects in railway track using the rail surface signal. Graz, Austria. (2022). |
dc.relation.references | W. Jeong, D. Jeong. Acoustic roughness measurement of railhead surface using an optimal sensor batch algorithm. Daejeon, South Korea. (2020). |
dc.relation.references | Z. Yuan, S. Zhu, C. Chang, X. Yuan, Q. Zhang. An unsupervised method based on convolutional variational auto-encoder and anomaly detection algorithms for light rail squat localization. Chengdu, China. (2021). |
dc.relation.references | M. Shadab, M. Ibrahim, I. Sarwar, H. Siddiqui, F. Rustam, E. Lee, I. Ashraf. Railway track inspection using deep learning based on audio to spectrogram conversion: an on-the-fly approach. Gyeongsan, South Korea. (2022). |
dc.relation.references | Y. Wang, P. Wang, Q. Wang, Z. Chen, Q. He. Using vehicle interior noise classification for monitoring urban rail transit infrastructure. Buffalo, USA. (2020). |
dc.relation.references | Integrated maintenance planning approach to optimize budget allocation for subway operating systems. Giza, Egypt. (2022). |
dc.relation.references | A.P De Man. Pin-pin resonance as a reference in determining ballasted railway track vibration behavior. Delft, Netherlands. (2000). |
dc.relation.references | R. Lewis, P. Christoforou, W.J. Wang, A. Beagles, M. Burstow, S.R. Lewis. Investigation of the influence of rail hardness on the wear of rail and wheel materials under dry conditions (ICRI wear mapping project). Sheffield, United Kingdom. (2019). |
dc.relation.references | E. Tutumluer, T.D. Stark, D. Mishra. Investigation and mitigation of differential movement at railway transitions for us high speed passenger rail and joint passenger/freight corridors. Pennsylvania, USA. (2012). |
dc.relation.references | Z. Li, A. Nuñez, M. Molodova, R. Dollevoet. Improvements in axle box acceleration measurements for the detection of light squats in railway infrastructure. Delft, Netherlands. (2015). |
dc.relation.references | Zhiling Guo, Qi Chen 1,2, Guangming Wu , Yongwei Xu, Ryosuke Shibasaki and Xiaowei Shao. Village building identification based on ensemble convolutional neural networks. Wuhan, China. (2017). |
dc.rights.accessrights | info:eu-repo/semantics/openAccess |
dc.subject.lemb | Rieles (Ferrocarriles) |
dc.subject.lemb | Ferrocarriles - Mantenimiento y reparación |
dc.subject.lemb | Vías férreas - Mantenimiento y reparación |
dc.subject.lemb | Transporte ferroviario |
dc.subject.lemb | Análisis de Fourier |
dc.subject.lemb | Dinámica de estructuras |
dc.subject.lemb | Desgaste mecánico |
dc.subject.proposal | Función de respuesta de frecuencia (FRF) |
dc.subject.proposal | Método de elementos finitos (MEF) |
dc.subject.proposal | Frecuencia de fijaciones |
dc.subject.proposal | Transformada de Fourier |
dc.subject.proposal | Test de martillo |
dc.subject.proposal | Red Neuronal Profunda |
dc.subject.proposal | Transformada de Wavelet |
dc.subject.proposal | Ensayo de martillo de impacto |
dc.subject.proposal | Frecuencia pin-pin |
dc.subject.proposal | Frequency response function (FRF) |
dc.subject.proposal | Finite element method |
dc.subject.proposal | Pin-Pin frequency |
dc.subject.proposal | Fourier Transform |
dc.subject.proposal | Hammer test |
dc.subject.proposal | Deep Neural Networks |
dc.subject.proposal | Wavelet Transform |
dc.subject.proposal | Impact Hammer Test |
dc.title.translated | Characterizing the condition of railroad track by using Frequency Response Function and Deep Neural Networks |
dc.type.coar | http://purl.org/coar/resource_type/c_bdcc |
dc.type.coarversion | http://purl.org/coar/version/c_ab4af688f83e57aa |
dc.type.content | Text |
dc.type.redcol | http://purl.org/redcol/resource_type/TM |
oaire.accessrights | http://purl.org/coar/access_right/c_abf2 |
dcterms.audience.professionaldevelopment | Investigadores |
dcterms.audience.professionaldevelopment | Público general |
dc.description.curriculararea | Ingeniería Mecánica.Sede Medellín |
Archivos en el documento
Este documento aparece en la(s) siguiente(s) colección(ones)
Esta obra está bajo licencia internacional Creative Commons Reconocimiento-NoComercial 4.0.Este documento ha sido depositado por parte de el(los) autor(es) bajo la siguiente constancia de depósito