Improved realism for heating

[Swamp-Ig]

At the moment the heat equation isn't very realistic, and gives some weird results.

This bug is to track improvements.

The model I am envisaging has three thermal components, the plasma, the tiles, and the part (where the part temp interacts with the rest of KSP)

Definition of terms

T_plasma The temperature of the plasma
T_tiles Temp tiles
T_part Part the part temp, what KSP ends up seeing.
T_Atm Atmospheric temp.

ρ_Atm Atmospheric density ∝ Atmospheric pressure
v Velocity of re-entering capsule
A Frontal area of capsule.
t Time.

To make it easy, I'll only deal with fluxes (heat change per unit area), Since heatshield mass and area are equivalent

T_plasma

This is going to be some function of T_Atm, v_capsule, ρ_Atm.
Will need to do some lit searching to figure it out. I guess A is likely to be another dependent variable.

Thermal coupling between plasma and tiles

This will be made up of convective and radiative transfer. At lower temps, the convective transfer dominates and is ∝ the difference between the plasma and tile temps (ΔT). At higher temps radiative transfer begins to dominate. Radiative intensity is ∝ ΔT⁴.

There's no real heat loss to the plasma since it's constantly renewed. So final equation is:

dT_tiles/dt = A ΔT + B ΔT⁴

Ablation

Ablation can be modelled as a function of T_tiles. Ablation will 'take away' some of the heat from the thermal coupling with the plasma, but that can just be pushed back to

tbc