Date of Award
Doctor of Philosophy
Aly E. Fathy
Syed K. Islam, Gong Gu, Thomas T. Meek
Having access to powerful processors allows scientists to carry out aggressive numerical computations to bridge the gaps which already exist among different fields of physics by exploring new multi-physics models to approach real life models of various phenomena happening around us in real life and accounting of the various coupling and dependence between the various physical parameters and material parameters.
Scientists greatly appreciate multi-physics modeling as they recognize:
1- Prototyping is expensive
2- Most of available CAD tools are not addressing the real model or accounting between the different physical parameters
3- Some difficulties to optimize the real model without simulation (due to a wide range of parameters)
4- Even some measurements are complicated and require expensive set-ups for full evaluation and trouble shooting. Some parameters are even non-measurable.
Solving multi-physics problems is always a complicated process. For instance, access to a reliable database to initialize data or to set boundary conditions is an important issue, but it may take long time (several years) to extract data from measurements. Nowadays, there is a big competition among commercial CAD tools developers to provide specific features or capabilities (as a multi-physics solvers). Fortunately, by developing powerful CAD tools, some complex questions will be answered, and also it may turn out to lead to some exciting innovations. In this study, two relevant electrical engineering problems (improving solar cells’ efficiency and effect of heat on electronic circuits like processes) will be investigated. The first one is the analysis of solar cells, and the second is speed deterioration of microprocessors due to generated heat. Both problems require to do some multi-physics modeling, and to solve them it requires investigating different physical set of parameters. Solar cell’s operation includes optics, semiconductor physics, and electronics. Modeling a solar cell could turn out to achieve higher efficiency. Similarly, solving a microprocessor’s problems include heat analysis, and mechanical stress analysis. Modeling microprocessors can lead to better heat sinks or cooling and could lead to sustained speeds and performance.
Ghahremani, Ahmadreza, "MULTI-PHYSICS MODELING. " PhD diss., University of Tennessee, 2016.