Thermal system analyses are often required, and conducted, in engineering problems in almost all engineering disciplines. The methods employed vary a great deal and must be specially adapted for each situation. Basically, two different methods are used, namely distributed modelling and lumped parameter modelling.

In this thesis, the methods are employed in two projects on applied engineering heat-transfer and temperature calculations. The two projects, although similar in many ways, show differences in ways that make a methodology differentiation necessary. In both projects the time-dependent temperature of the system is studied, and the principal method is combined with additional, complementary, and experimental methods. Both projects have industrial applications.

The first project discusses heat-transfer considerations in automated glass manufacturing. The manufacturing process consists of several steps; one step being of special interest for the final production result, namely the forming of a parison-shaped glass blank. The blank is pressed from a glass gob and will in the step following the pressing be blown to a bottle. The temperature distribution in the parison at the end of the pressing is decisive for the glass distribution in the bottle. The aim of the study is to obtain a better knowledge of the temperature distribution in the parison and a better understanding of the decisive heat-transfer mechanisms.

The time-dependent temperature distribution in the parison and in the manufacturing tools is analysed. The heat-transfer problem is complicated by the internal radiation in the semitransparent glass and the time-dependent contact resistance to heat transfer between the glass and the tools.

The second project is on thermal and thermodynamic considerations in fluid temperature- control systems. The temperature control of the fluid is important for a number of reasons; the fluid temperature itself may be of importance or may have secondary implications, e.g. fluid degeneration, lubrication problems or other temperature-related faults. The aim is to predict the time-dependent fluid temperature throughout the system and to obtain a better understanding of the influencing thermal and thermodynamic mechanisms. The project covers the analysis of a thermostatic mixing valve, where large fluctuations in the outlet water temperature are experienced, and a fluid power system, where oil cooling is required in order to avoid highly elevated oil temperatures. In both cases the fluid temperature is controlled by thermostatic control circuits.

The project is a lumped parameter analysis of thermal changes and variations in relatively large systems, consisting of a number of different components and subsystems. Whereas a lumped formulation is used, distributed modelling of components is often required as a complement in order to ensure accurate system modelling.