Thesis Modelo de distribución de calor y control térmico de mesa de calor mediante sensores de fibra óptica
Loading...
Date
2025-08
Authors
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Civil Electrónica
Departament
Campus
Campus Casa Central Valparaíso
Abstract
Este trabajo aborda el modelado, estimación y control de la temperatura en una mesa térmica de cobre instrumentada con sensores de fibra óptica basados en rejillas de Bragg(FBG). El objetivo principal es desarrollar un sistema capaz de predecir y regular el comportamiento térmico en distintos puntos de la superficie, combinando modelado numérico, observación de estados y control óptimo en tiempo discreto. El modelado térmico se realizó mediante el Método de Elementos Finitos (FEM) aplicado a la ecuación de difusión de calor bidimensional, considerando pérdidas por convección y excitación térmica localizada mediante un actuador tipo PTC. Los parámetros físicos del modelo se ajustaron mediante comparación con mediciones experimentales obtenidas a través de los sensores FBG, garantizando una representación coherente del fenómeno real. Sobre el modelo FEM discretizado se construyó un modelo reducido mediante balanced truncation, conservando las dinámicas dominantes. Este modelo sirvió de base para implementar un observador discreto de Luenberger, diseñado con la técnica place, capaz de estimar temperaturas en zonas no sensadas directamente. La validación del observador arrojó errores promedio menores a 1 °C entre mediciones y estimaciones, confirmando su precisión tanto en sensores de corrección como de validación independiente. Posteriormente, se diseñó un controlador LQI discreto(...).
This work addresses the modeling, estimation, and control of temperature in a copper thermal plate instrumented with Fiber Bragg Grating (FBG) optical sensors. The main objective is to develop a system capable of predicting and regulating the thermal behavior at different surface points by combining numerical modeling, state observation, and optimal discrete-time control. The thermal modeling was performed using the Finite Element Method (FEM) applied to the two-dimensional heat diffusion equation, considering convective losses and localized thermal excitation through a PTC-type actuator. The physical parameters of the model were tuned through comparison with experimental measurements obtained via FBG sensors, ensuring a consistent representation of the real thermal behavior. Based on the discretized FEM model, a reduced-order model was constructed using balanced truncation, preserving the dominant dynamics. This model served as the foundation for implementing a discrete Luenberger observer, designed via the place method, capable of estimating temperatures in regions not directly measured. The validation of the observer yielded average errors below 1 °C between measured and estimated temperatures, confirming its accuracy both in correction and independent validation sensors. Subsequently, a discrete LQI controller was designed(...).
This work addresses the modeling, estimation, and control of temperature in a copper thermal plate instrumented with Fiber Bragg Grating (FBG) optical sensors. The main objective is to develop a system capable of predicting and regulating the thermal behavior at different surface points by combining numerical modeling, state observation, and optimal discrete-time control. The thermal modeling was performed using the Finite Element Method (FEM) applied to the two-dimensional heat diffusion equation, considering convective losses and localized thermal excitation through a PTC-type actuator. The physical parameters of the model were tuned through comparison with experimental measurements obtained via FBG sensors, ensuring a consistent representation of the real thermal behavior. Based on the discretized FEM model, a reduced-order model was constructed using balanced truncation, preserving the dominant dynamics. This model served as the foundation for implementing a discrete Luenberger observer, designed via the place method, capable of estimating temperatures in regions not directly measured. The validation of the observer yielded average errors below 1 °C between measured and estimated temperatures, confirming its accuracy both in correction and independent validation sensors. Subsequently, a discrete LQI controller was designed(...).
Description
Keywords
Control de Temperatura, Observador de Luenberger, Control óptimo discreto (LQI), Método de elementos finitos (FEM), Rejillas de bragg (FBG), Ingeniería de control y automatización
