Convection is a mechanism of heat transport in which moving particles absorb thermal energy at one location and release it at another. It is thus bound to a mass transport that cannot take place under conditions of vacuum or in a solid, but is almost unavoidable in liquids and gases. Convection therefore depends on the mobility of the particles involved and is therefore firmly coupled to the flow conditions in the fluid.
For the prediction of temperature fields in structures in which convection is involved, the convective heat transfer or the heat transfer coefficient on the wall between solid and fluid is of particular interest. It corresponds to a superposition of the two physical mechanisms, heat conduction and convection. The heat transfer coefficient itself is defined by the amount of heat that flows through a wall when there is a temperature difference between the surface and the fluid. The size of the heat transfer coefficient therefore depends on the material properties, the physical state and the flow conditions close to the wall.
Depending on the underlying cause of the flow, a distinction is made between free and forced convection. With free or natural convection, the particle movement is caused by the temperature-related density difference in the fluid, whereas with forced convection, external forces are responsible for this.
Computation time / model size
Depending on the required or desired result quality, the heat transfer due to convection must be represented in the simulation with different models. If a good approximation is sufficient, the heat transfer coefficient and the associated fluid temperature can be specified as boundary conditions in a simple temperature simulation. As a general rule, larger areas with similar conditions are combined. If, however, the exact spatial distribution of the heat transfer coefficients is necessary or the flow conditions show a strong dependence on the wall temperature and/or the heat flow, a coupled flow and temperature simulation is required.
The difference in the effort involved in modeling and solving the systems of equations is correspondingly large. In many cases, however, a simple temperature field simulation is sufficient. Since in this case only the temperature is present as a degree of freedom, a relatively small system of equations is obtained, which can be solved relatively quickly. This approach is most efficient, especially in the concept phase, when it is less about absolute values than about a fundamental evaluation of different design variants.
A temperature field simulation at which no convection takes place is reserved for special cases. Their importance for thermal simulations is correspondingly high.