Numerical study of transient heat transfer in a rotating roll exposed to heat flux and convection
The hot rolling process experiences intense thermal cycles: they receive peaks of heat flux when the steel strip contacts them and, immediately afterwards, are cooled by water sprays or by the environment. If the temperature gradient is not controlled, cracks and ovalizations appear that shorten the...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/446071 |
| Acceso en línea: | https://hdl.handle.net/2117/446071 |
| Access Level: | acceso abierto |
| Palabra clave: | Thermodynamics Fluid mechanics Strength of materials Termodinàmica Mecànica de fluids Resistència de materials Àrees temàtiques de la UPC::Enginyeria mecànica |
| Sumario: | The hot rolling process experiences intense thermal cycles: they receive peaks of heat flux when the steel strip contacts them and, immediately afterwards, are cooled by water sprays or by the environment. If the temperature gradient is not controlled, cracks and ovalizations appear that shorten the operating life of the roll and generate surface defects in the strip. The main objective of this Master’s Thesis is to develop a numerical tool that anticipates, in real time, the transient evo- lution of temperature within a solid rotating roll, thereby facilitating the design and control of cooling systems. To this end, an existing analytical solution is used as a reference and three numerical methods in a transient regime are implemented: (i) a first-order explicit scheme, simple but limited by stability; (ii) a classical implicit Crank–Nicolson scheme, which exhibits high accuracy; and (iii) a second-order implicit scheme (BDF2), which combines robustness and speed. The algorithms are programmed in Python. The evaluation compares the accuracy and computational cost of the three methods against the reference analytical solution, with the aim of identifying the most convenient procedure according to industrial criteria of allowable error and computation time. The work concludes with the advantages of each method and describes the following steps of the study, necessary before performing extensions to refined meshes and more complex assemblies. |
|---|