# IEA-ETSAP Forum

Full Version: Seasonal HP
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Hi, I am modelling a HP where I want to take into consideration the changes of performance through the different outdoor temperatures during the year. For that, three process are created, one to take the investment cost into consideration and two other processes, one to account for low efficiency in winter and another to account for high efficiency in Summer. Then, AF is used to use the process with high efficiency in summer (but not in winter) and the process with low efficiency in winter (but not in summer). So these three processes would represent, let's say, an electric heat pump. A commodity (ie: HPINV) is flowing out of the investment process and is going into the two other processes. other two commodities are also going into this other two processes, Ambient temperature (AHT) and Electricity (ELC). Question: How can I establish that the production of each unit of heat (RHUN) out of these two processes needs 0.25 units of ELC, 0.75 units of AHT.....PLUS one unit of HPINV?  (I think Share cannot be used because I do not want HPINV to be part of the share). Or do you have a best practice to model seasonal changes in HP performance? Thanks.
I don't quite understand the reason for three processes; why not model the seasonal efficiency differences in a single heat pump process, more like in real-life?  Availabilities, efficiencies, shares etc. attributes have the timeslice index just for that purpose. For a thread with similar issues see here: Heat Pumps -- Best practice
Anyway, just to answer the question:
Diana Wrote:How can I establish that the production of each unit of heat (RHUN) out of these two processes needs 0.25 units of ELC, 0.75 units of AHT.....PLUS one unit of HPINV?
That would be very easy. You can just define Input(ELC)=0.25, Input(AHT)=0.75 and Input(HPINV)=1. (Input is a VEDA attribute, alias for VDA_FLOP(r,y,p,cg,s), and so you can also differentiate it by season.) [EDIT:] In the above I was assuming that the process is defined on the DAYNITE or SEASON level, on the basis of the model you posted earlier. For users defining end-use heat-pumps on the ANNUAL level, note that in their case Input cannot be differentiated by timeslice (because it is defined from the activity to the input flow), but ACT_EFF and FLO_SHAR remain timeslice-dependent. Thus, when the process level is ANNUAL, you could define HPINV as an auxiliary commodity (or of type FIN), and then define e.g. ACT_EFF(NRG)=1, FLO_SHAR(ELC,AHT,FX)=0.25/0.75 = 0.333333, and Input(HPINV)=1. With different timeslice-dependent fixed FLO_SHAR values, you can then easily vary the ratio of ELC to AHT by season.
Dear Antti, Thanks for your help.