Unicron, the orbital duration of TRAPPIST-1 d is 4.05 days. The orbital duration of 18.202 days refers to the old TRAPPIST-1 version when only 3 planets were known. The data for TRAPPIST-1 d were inaccurate.
SpaceEngineer, in your TRAPPIST-1 pak under 0.9.8, TRAPPIST-1 b is an oceania with no clouds. (You can see this in the screenshot kham132 posted on page 2.) Is the lack of clouds unrealistic, and should the script have a cloud layer added only to that planet?
Actually, it is for SE version0.981. The 'PeriodDays' parameter does not exist until 0.981. By the way, this is the reason that in SE 0.980 as orbital time for TRAPPST-1 d is displayed 18,202 days. The parameter 'PeriodDays' is ignored, and the value is taken from the old script of TRAPPIST-1, that only contains three planets.
I fixed the addon, update link is in the OP. I also confused the planets classes: it's planet а must be oceania, not g, because of its density (3.3 g/cm3). Planet h is too dense to ice world, it must be a desert too. Now classes are:
b - oceania c - desert d - terra e - terra f - oceania g - desert h - desert
Planets d and e are terras to make the addon more fancy, but in reality they are most likely deserts.
Star luminosity is increased 2 times to make temperatures on planets more high, because in SE temperature calculation on planets of cool stars is somewhat inaccurate.
Wow nice! There is a last minor error in my addon: mass of the planet h is 0.41, it was copy-pasted from planet d. This is because I made the script by copy-pasting blocks of code... But anyway 0.41 Earth masses is a realistic mass for a planet of 0.75 Earth radii. If removing mass of the planet h from the script, SE generates mass of 0.2 for it, and making it an ice world.