Show simple item record

dc.rights.licenseAtribución-NoComercial-SinDerivadas 4.0 Internacional
dc.contributor.advisorMartínez Carvajal, Hernán Eduardo
dc.contributor.authorArcila Quintero, Norbey
dc.date.accessioned2023-01-17T21:49:22Z
dc.date.available2023-01-17T21:49:22Z
dc.date.issued2022-11
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/82998
dc.description.abstractSe realizó un modelo para predicción de niveles de alerta para fallas de geomateriales utilizando el método del Inverso de la Velocidad del desplazamiento en conjunto con métodos de aprendizaje de máquinas de tipo supervisado, que puede ser utilizado en planes de monitoreo geotécnico continuo. Para el modelo se propuso el uso de Máquinas de Soporte Vectorial (SVM), y Árboles de Decisión como métodos de aprendizaje de máquinas a evaluar, obteniendo resultados de desempeño con la matriz de confusión cercanos a 0,6 en el primero y de 0,98 con el segundo. En ambos casos el entrenamiento del modelo se ejecutó con datos sintéticos y la evaluación se efectuó empleando datos de casos reales donde ocurrió una falla del geomaterial monitoreado. El modelo demuestra la idoneidad de la utilización de datos sintéticos para su entrenamiento, especialmente para el modelo implementado con árboles de decisión. No obstante, es clara la dependencia del modelo a datos de instrumentación geotécnica de tipo automatizada para la aplicación del plan de monitoreo continuo. Esto para poder contar con un sistema de predicción en tiempo cuasi real, además de que se hace incuestionable que se obtiene un mejor desempeño cuando se cuentan con volúmenes de información superiores a 8 datos por hora, asimismo para la metodología desarrollada es imprescindible contar con un registro de datos con un espaciado de tiempo constante. (texto tomado de la fuente)
dc.description.abstractA model is made to predict alert levels for geomaterial failures using the inverse velocity method with supervised machine learning, which can be used in continuous geotechnical monitoring plans. For the model, it proposes the use of Support Vector Machines (SVM), and Decision Trees as machine learning methodologies to be evaluated, obtaining performance results with the confusion matrix close to 0.6 in the first and 0.98 with the second. In both cases, the model training was executed with theoretical data and the evaluation was performed using data from real cases where a failure of the monitored geomaterial occurred. The model demonstrates the suitability of using theoretical data for model training, especially for the methodology implemented with decision trees. However, the dependence of the model on automated geotechnical instrumentation data for the application of the continuous monitoring plan is evident, thus to have a prediction system in quasi real time. In addition, it is unquestionable to the developed methodology that better performance is XII Predicción de alertas de falla para planes de monitoreo geotécnico obtained when there are volumes of information greater than 8 data per hour. Plus, it is essential to have a data recording with a constant time interval.
dc.format.extentxix, 109 páginas
dc.format.mimetypeapplication/pdf
dc.language.isospa
dc.publisherUniversidad Nacional de Colombia
dc.rights.urihttp://creativecommons.org/licenses/by-nd/4.0/
dc.subject.ddc620 - Ingeniería y operaciones afines::624 - Ingeniería civil
dc.titlePredicción de alertas de falla para planes de monitoreo geotécnico aplicando el método del inverso de la velocidad acoplado con algoritmos de aprendizaje supervisado
dc.typeTrabajo de grado - Maestría
dc.type.driverinfo:eu-repo/semantics/masterThesis
dc.type.versioninfo:eu-repo/semantics/acceptedVersion
dc.publisher.programMedellín - Minas - Maestría en Ingeniería - Geotecnia
dc.description.notesContiene una tesis de maestría donde se desarrolla un modelo para la predicción de niveles de alerta de fallas en geomateriales, usando el método del Inverso de la Velocidad del desplazamiento implementado con algoritmos de aprendizaje supervisado, que pueda ser aplicable de manera continua a los planes de monitoreo de instrumentación geotécnica.
dc.contributor.researchgroupGrupo de Geotecnia
dc.description.degreelevelMaestría
dc.description.degreenameMaestría en Ingeniería - Geotecnia
dc.description.researchareaInstrumentación Geotécnica
dc.identifier.instnameUniversidad Nacional de Colombia
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombia
dc.identifier.repourlhttps://repositorio.unal.edu.co/
dc.publisher.facultyFacultad de Minas
dc.publisher.placeMedellín, Colombia
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellín
dc.relation.indexedLaReferencia
dc.relation.referencesA. Federico, M. Popescu, G. Elia, C. Fidelibus, G. Interno y A. Murianni, «Prediction of time to slope failure: A general framework,» Environmental Earth Sciences, vol. 66, p. 245–256, 2012.
dc.relation.referencesE. Intrieri, T. Carlà y G. Gigli, «Forecasting the time of failure of landslides at slope-scale: A literature review,» Earth-Science Reviews, vol. 193, pp. 333-349, 2019.
dc.relation.referencesK. Terzaghi, «Mechanism of Landslides,» de Application of Geology to Engineering Practice, Geological Society of America, 1950, p. 83–125.
dc.relation.referencesD. J. Varnes, «Time-deformation relations in creep to failure of earth materials,» de 7th Southeast Asia Geotechnical Conference, Hong Kong, 1982.
dc.relation.referencesF. Tavenas y S. Leroueil, «Creep and failure of slopes in clays,» Canadian Geotechnical Journal, vol. 18, nº 1, p. 106–120, 1981.
dc.relation.referencesT. Fukuzono, «A new method for predicting the failure time of slopes,» de 4th International Conference & Field Workshop on Landslides, Tokyo, 1985.
dc.relation.referencesA. Manconi y D. Giordan, «Landslide failure forecast in near-real-time,» Geomatics, Natural Hazards and Risk, vol. 7, nº 2, pp. 639 - 648, 2016.
dc.relation.referencesT. Carlà, E. Intrieri, F. Di Traglia, T. Nolesini, G. Gigli y C. N, «Guidelines on the use of inverse velocity method as a tool for setting alarm thresholds and forecasting landslides and structure collapses,» Landslides, vol. 14, p. 517 – 534, 2017.
dc.relation.referencesA. Segalini, A. Valletta y A. Carri, «Landslide time-of-failure forecast and alert threshold assessment: A generalized criterion,» Engineering Geology, vol. 245, pp. 72-80, 2018.
dc.relation.referencesE. Intrieri, G. Gigli, F. Mugnai, . R. Fanti y N. Casagli, «Design and implementation of a landslide early warning system,» Engineering Geology, Vols. %1 de %2147-148, pp. 124-136, 2012.
dc.relation.referencesA. Baghbani, . T. Choudhury, S. Costa y J. Reiner, «Application of artificial intelligence in geotechnical engineering: A state-of-the-art review,» Earth-Science Reviews, vol. 228, 2022.
dc.relation.referencesF. Marconi, Newsmakers: Artificial Intelligence and the Future of Journalism, Nueva York: Columbia University Press, 2020.
dc.relation.referencesS. C. Jong, D. E. Ong y E. Oh, «State-of-the-art review of geotechnical-driven artificial intelligence techniques in underground soil-structure interaction,» Tunnelling and Underground Space Technology incorporating Trenchless Technology Research, vol. 113, 2021.
dc.relation.referencesM. Jäggli, «Il cataclisma tellurico al Motto d'Arbino,» Ticino , vol. 5, nº 11, pp. 161-164, 1928.
dc.relation.referencesA. Heim, «Bergsturz und Menschenleben,» p. 218, 1932.
dc.relation.referencesM. Saito y M. Uezawa, «Failure of soil due to creep,» de the 5th International Conference on Soil Mechanics and Foundation Engineering, Mexico, 1961.
dc.relation.referencesM. Saito, «Forecasting the Time of Occurrence of a Slope Failure,» de 6th International Conference on Soil Mechanics and Foundation Engineering, Montreal, 1965.
dc.relation.referencesM. Saito, « Forecasting Time of Slope failure by Tertiary Creep,» de the 7th International Conference on Soil Mechanics and Foundation Engineering, Mexico, 1969.
dc.relation.referencesG. J. Dick , E. Eberhardt, A. G. Cabrejo-Liévano, D. Stead y N. D. Rose, «Development of an early-warning time-of-failure analysis methodology for open-pit mine slopes utilizing ground-based slope stability radar monitoring data,» Canadian Geotechnical Journal, vol. 52, nº 4, pp. 515-529, 2014.
dc.relation.referencesJ. N. Hutchinson, « Landslide risk—to know, to foresee, to prevent,» Geol Tecnica e Ambientale, vol. 9, pp. 3-24, 2001.
dc.relation.referencesO. Hungr y A. Kent, «Coal mine waste dump failures in British Columbia, Canada.,» Landslide News, vol. 9, pp. 26-28, 1995.
dc.relation.referencesN. D. Rose ND y O. Hungr, «Forecasting potential rock slope failure in open pit mines using the inverse-velocity method,» International Journal of Rock Mechanics and Mining Sciences, vol. 44, nº 2, pp. 308-320, 2007.
dc.relation.referencesT. Fukuzono, «Recent studies on time prediction of slope failure.,» Landslide News, vol. 4, pp. 9-12, 1990.
dc.relation.referencesB. Voight, «A Relation to Describe Rate-Dependent Material Failure,» Science, vol. 243, nº 4888, pp. 200-203, 1989.
dc.relation.referencesB. Voight, «Materials science law applies to time forecasts of slope failure,» Landslide News, vol. 3, pp. 8-11, 1989.
dc.relation.referencesG. B. Crosta y F. Agliardi, «Failure forecast for large rock slides by surface displacement measurements,» Canadian Geotechnical Journal, vol. 40, p. 176 – 191, 2003.
dc.relation.referencesT. Jo, Machine Learning Foundations, Gewerbestrasse: Springer, 2021.
dc.relation.referencesM. Awad y R. Khanna, Efficient Learning Machines: Theories, Concepts, and Applications for Engineers and System Designers, New York: Apress Open, 2015.
dc.relation.referencesV. Vapnik, The Nature of Statistical Learning Theory, New York: Springer, 1995.
dc.relation.referencesA. B. Andre, E. Beltrame y J. Wainer, « A combination of support vector machine and k-nearest neighbors for machine fault detection,» Applied Artificial Intelligence, vol. 27, pp. 36-49, 2013.
dc.relation.referencesJ. A. Rodrigo, «Cienciadedatos.net,» Abril 2017. [En línea]. Available: https://www.cienciadedatos.net/documentos/34_maquinas_de_vector_soporte_support_vector_machines. [Último acceso: 20 julio 2022].
dc.relation.referencesJ. Orellana Alvear, «Arboles de decision y Random Forest,» 12-16 Noviembre 2018. [En línea]. Available: https://bookdown.org/content/2031/arboles-de-decision-parte-i.html. [Último acceso: 17 Julio 2022].
dc.relation.referencesIBM, «Modelos de árboles de decisión,» 17 08 2021. [En línea]. Available: https://www.ibm.com/docs/es/spss-modeler/saas?topic=trees-decision-tree-models. [Último acceso: 25 06 2022].
dc.relation.referencesHuawei, «Método Supervisado - Random Forests,» 12 05 2022. [En línea]. Available: https://forum.huawei.com/enterprise/es/m%C3%A9todo-supervisado-random-forests/thread/873247-100757. [Último acceso: 20 07 2022].
dc.relation.referencesIBM, «IBM Cloud Education,» 3 03 2021. [En línea]. Available: https://www.ibm.com/cloud/learn/overfitting. [Último acceso: 29 06 2022].
dc.relation.referencesF. Bozzano, I. Cipriani, P. Mazzanti y A. Prestininzi , «A field experiment for calibrating landslide time-of-failure prediction functions,» International Journal of Rock Mechanics & Mining Sciences, vol. 67, pp. 69-77, 2014.
dc.relation.referencesT. Carlà, P. Farina, E. Intrieri y N. Casagli, «Integration of ground-based radar and satellite InSAR data for the analysis of an unexpected slope failure in an open-pit mine,» Engineering Geology, vol. 235, pp. 39-52, 2018.
dc.relation.referencesJ. D. Graham, DEVELOPMENT OF AN EARLY WARNING TIME-OF-FAILURE ANALYSIS METHODOLOGY FOR OPEN PIT MINE SLOPES UTILIZING THE SPATIAL DISTRIBUTION OF GROUND-BASED RADAR MONITORING DATA, Vancouver, 2013.
dc.relation.referencesMATLAB, «Suavizado de datos y detección de valores atípicos,» Mathworks, 2021. [En línea]. Available: https://es.mathworks.com/help/matlab/data_analysis/data-smoothing-and-outlier-detection.html. [Último acceso: 29 07 2022].
dc.relation.referencesEsri, «Cómo funciona Suavizado de serie temporal,» ArcgisPro, 2021. [En línea]. Available: https://pro.arcgis.com/es/pro-app/2.8/tool-reference/spatial-statistics/how-time-series-smoothing-works.htm. [Último acceso: 26 07 2022].
dc.relation.referencesJ. C. González-Avella, J. M. Tudurí y G. Rul-lan, «apsl.net,» 14 junio 2017. [En línea]. Available: https://www.apsl.net/blog/2017/06/14/analisis-de-series-temporales-usando-redes-neuronales-recurrentes/. [Último acceso: 29 07 2022].
dc.relation.referencesT. Carlà, R. Macciotta, M. Hendry, D. Martin, T. Edwards, T. Evans, P. Farina, E. Intrieri y N. Casagli, «Displacement of a landslide retaining wall and application of an enhanced failure forecasting approach,» Landslides, vol. 15, p. 489–505, 2017.
dc.relation.referencesR. Fernandes de Mello y M. Antonelli Ponti, Machine Learning: A Practical Approach on the Statistical Learning Theory, Springer Cham, 2018.
dc.rights.accessrightsinfo:eu-repo/semantics/openAccess
dc.subject.lembGeotectónica
dc.subject.proposalDatos sintéticos
dc.subject.proposalInstrumentación geotécnica
dc.subject.proposalInstrumentación Geotécnica
dc.subject.proposalDatos sintéticos
dc.title.translatedPrediction of fault alerts for geotechnical monitoring plans applying the inverse velocity method coupled with supervised learning algorithms.
dc.type.coarhttp://purl.org/coar/resource_type/c_bdcc
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.contentText
dc.type.redcolhttp://purl.org/redcol/resource_type/TM
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2
dc.description.curricularareaÁrea Curricular de Ingeniería Civil
dc.contributor.orcid0000-0001-6963-2741


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Atribución-NoComercial-SinDerivadas 4.0 InternacionalThis work is licensed under a Creative Commons Reconocimiento-NoComercial 4.0.This document has been deposited by the author (s) under the following certificate of deposit