Evaluación de una propuesta de instrumentación para shiploader mediante prototipo a escala y modelación numérica.
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
El transporte de minerales en puertos es clave para la industria minera, y los shiploaders desempeñan un papel fundamental en este proceso. Sin embargo, estos equipos enfrentan cargas dinámicas y condiciones operativas que pueden causar daños estructurales con el tiempo. Este trabajo tiene como objetivo evaluar la efectividad de un sistema de instrumentación propuesto para detectar daños en un shiploader, utilizando un modelo a escala instrumentado y simulación numérica. Los objetivos incluyen desarrollar un prototipo a escala, instrumentarlo con sensores, y validar su comportamiento mediante modelación con elementos finitos (MEF).
En este estudio se utilizan dos enfoques de detección de daño: (1) un enfoque basado en física y (2) un enfoque basado en los datos. La metodología se basó en la construcción de un prototipo a escala con piezas Meccano, adaptado a partir de un diseño conceptual, e instrumentado con strain gauges, que registran deformaciones, y un acelerómetro, que se utiliza como inclinómetro para informar la inclinación del brazo. Se desarrolló un modelo numérico en SAP2000, que replica la condición operativa y permite comparar los resultados experimentales con los simulados. Para evaluar la efectividad de la instrumentación, se definieron tres escenarios de daño estructural simulados, los cuales son el debilitamiento en las diagonales y aflojamiento de conexiones, e impactos en el brazo y la torre. Los datos experimentales fueron procesados aplicando factores de conversión y la Transformada Rápida de Fourier (FFT) para eliminar ruido. Se utilizaron modelos en Autodesk Inventor para interpretar las lecturas de deformación en términos de fuerza y momento flector.
Los resultados mostraron que, si bien los sensores detectaron daños simulados, la propuesta de instrumentación es mejorable ya que no considera instrumentar las zonas de mayor esfuerzo. El enfoque basado en los datos demostró ser más fiable y aplicable que el enfoque basado en la física, debido a las limitaciones propias de modelos numéricos. Se concluye que, aunque el modelo físico escalado presentó deficiencias, los datos obtenidos fueron valiosos para entender el comportamiento estructural de la estructura real. Se recomienda ampliar la instrumentación tal que se pueda detectar y desagregar la acción de fuerza axial y flexión fuera del plano en miembros críticos del shiploader.
The transportation of minerals at ports is key for the mining industry, and shiploaders play a fundamental role in this process. However, these machines face dynamic loads and operational conditions that can cause structural damage over time. The objective of this work is to evaluate the effectiveness of a proposed instrumentation system for detecting damage in a shiploader, using an instrumented scale model and numerical simulation. The goals include developing a scale prototype, instrumenting it with sensors, and validating its behavior through finite element modeling (FEM). This study uses two damage detection approaches: (1) a physics-based approach and (2) a data-driven approach. The methodology was based on constructing a scale prototype using Meccano parts, adapted from a conceptual design, and instrumented with strain gauges to measure deformations, and an accelerometer used as an inclinometer to report the tilt of the arm. A numerical model was developed in SAP2000, which replicates the operational condition and allows comparing experimental results with simulated ones. To assess the effectiveness of the instrumentation, three simulated structural damage scenarios were defined: weakening of the diagonals and loosening of connections, as well as impacts on the arm and tower. The experimental data were processed using conversion factors and the Fast Fourier Transform (FFT) to remove noise. Models in Autodesk Inventor were used to interpret the deformation readings in terms of force and bending moment. The results showed that, while the sensors detected simulated damage, the proposed instrumentation can be improved, as it does not consider instrumenting areas of higher stress. The data-driven approach proved to be more reliable and applicable than the physics-based approach due to the limitations of numerical models. It is concluded that, although the scaled physical model had shortcomings, the obtained data were valuable for understanding the structural behavior of the actual structure. It is recommended to expand the instrumentation so that axial force and out-of-plane bending actions can be detected and broken down in critical members of the shiploader.
The transportation of minerals at ports is key for the mining industry, and shiploaders play a fundamental role in this process. However, these machines face dynamic loads and operational conditions that can cause structural damage over time. The objective of this work is to evaluate the effectiveness of a proposed instrumentation system for detecting damage in a shiploader, using an instrumented scale model and numerical simulation. The goals include developing a scale prototype, instrumenting it with sensors, and validating its behavior through finite element modeling (FEM). This study uses two damage detection approaches: (1) a physics-based approach and (2) a data-driven approach. The methodology was based on constructing a scale prototype using Meccano parts, adapted from a conceptual design, and instrumented with strain gauges to measure deformations, and an accelerometer used as an inclinometer to report the tilt of the arm. A numerical model was developed in SAP2000, which replicates the operational condition and allows comparing experimental results with simulated ones. To assess the effectiveness of the instrumentation, three simulated structural damage scenarios were defined: weakening of the diagonals and loosening of connections, as well as impacts on the arm and tower. The experimental data were processed using conversion factors and the Fast Fourier Transform (FFT) to remove noise. Models in Autodesk Inventor were used to interpret the deformation readings in terms of force and bending moment. The results showed that, while the sensors detected simulated damage, the proposed instrumentation can be improved, as it does not consider instrumenting areas of higher stress. The data-driven approach proved to be more reliable and applicable than the physics-based approach due to the limitations of numerical models. It is concluded that, although the scaled physical model had shortcomings, the obtained data were valuable for understanding the structural behavior of the actual structure. It is recommended to expand the instrumentation so that axial force and out-of-plane bending actions can be detected and broken down in critical members of the shiploader.
Description
Tesis presentada para optar al título de Ingeniero/a Civil.
Keywords
Maquinaria en la industria, Industria minera, Mantenibilidad (Ingeniería)