The required heat pump capacity per building is calculated using the buildings peak heat demand. Due to some very high peaks this results in very high heat pump capacities, that are not very realistic for the buildings annual heat demand.
It is therefore suggested to cut off the peak at e.g. the 99,5 percentile. A small analysis for 15 MV grids showed, that cutting off the peaks at the 99,5 percentile would reduce the buildings annual heat demand by max. 1 %, while the heat pump capacity is reduced significantly (around 25 %).
The following needs to be done:
[ ] Implement cutting off the peak in the buildings heat demand time series here
[ ] Add percentile at which is clipped to config data
[ ] @ClaraBuettner should heat demand time series for eTraGo be calculated from the clipped time series data or from the unclipped? (Should only have impact on annual demand, not on the peaks)
[ ] @ClaraBuettner There is a table for the heat peak demand of each building. Would you
[ ] The reduced heat pump capacity will lead to infeasibilities in eDisGo's optimisation, as the heat pump capacity won't be sufficient to serve the peak heat demand. An issue needs to be made in eDisGo to fix that. One way could be to also clip the heat demand when it is retrieved from the database. The clipping could use the heat peak demand from the database to make sure, that it is consistent. Another option could be to add a storage per default, to buffer the difference. The storage could represent the building's thermal inertia.
The required heat pump capacity per building is calculated using the buildings peak heat demand. Due to some very high peaks this results in very high heat pump capacities, that are not very realistic for the buildings annual heat demand.
It is therefore suggested to cut off the peak at e.g. the 99,5 percentile. A small analysis for 15 MV grids showed, that cutting off the peaks at the 99,5 percentile would reduce the buildings annual heat demand by max. 1 %, while the heat pump capacity is reduced significantly (around 25 %).
The following needs to be done: