SurfaceControl: Movable Insulation vs Construction Properties @ EnergyPlus

edit

I would not expect the two to have the exact same behavior, because the conduction transfer functions will come out quite different. The heat transfer modeling for movable insulation is simplistic and affects the surface heat transfer, sort of like a film coefficient, while still using the one set of CTF coefficients for the main construction under the insulation. But when you include the insulation in the main construction you will get a different CTF and no tricks at the boundary condition. It is hard to speculate why your heating load changed the way it did but the two cases could easily have different impact on the coupling of thermal mass to the zone air. You might get closer by using Material:NoMass in the Construction experiment.

Another way to model dynamic materials is with EMS. You can vary the entire construction used for a surface over time with your own control program -- see the actuator Surface Construction State and the input object EnergyManagementSystem:ConstructionIndexVariable.

From my experience, as Archmage stated, you will have more accurate results with the "movableinsulation" model, in which no mass storage in the insulation layer is modelled, although this could be quite accurate (depending on the density of the insulation considered).

Moreover regarding the use of EMS and the object "EnergyManagementSystem:ConstructionIndexVariable", you can use this if the number of nodes, calculated by the CTF method, in the different constructions you are using remain the same for all the constructions. This is not usually the case if you change the thermal resistance of your construction. Therefore EnergyPlus will give you an error.

Please let me know if you have another experience with the EMS object. Fabio (ff279@cam.ac.uk)