jrissman
commented
3 years ago Adding a note here that as part of this update, the building energy efficiency standard policy (which is currently subscripted by building type and by building component) should also be subscripted by building fuel.
This will be the first policy lever with three subscripts, which currently cannot be handled by our web app. We hope to have a new web UI solution in place sometime in 2021 that can better handle extensively-subscripted policies, likely moving subscript elements out of the policy tree and into a subscript setting pane called up within each policy's pane.
(For a future EPS update)
Update the building sector structure to break apart floor area, service demand, equipment, and energy demand.
Challenges aside, this would allow for much better handling of building energy consumption and associate policies.
This update would require several new ways of disaggregating data:
1) Existing and new floor area. Adding in floor area and dis-aggregating by type would let us better handle retrofitting and building code policies. It might also enable modeling net zero energy building policies.
2) To calculate service demand, we can estimate the service demand energy use intensity of each floor area type. There is a fair amount of published literature to this effect. Also, to find the existing service demand intensity in the start year, we would multiply each piece of equipment's energy demand by the efficiency of that piece of equipment, to get the total service demand.
3) Use a decision mechanism for allocating new service demand to certain technologies. A very commonly used function is the logit function, which is easily parameterized and can be calibrated to historical data to achieve the correct shares of sales in the start year. This is what many of the existing IAM models do. We may deviate from projections in future years, but perhaps that is okay as long as the switch isn't dramatic. Dramatic changes can also be controlled through the exponent size and the share-weights
4) (Optional but recommended) track the equipment stock more precisely with survival functions (like the issue submitted for the transportation sector).
If we were to make these changes, we would have much better representation of building energy and policy impacts, and we could much more easily model financial incentives and rebates. We also could much more easily model building capital costs because we would collect the appliance cost data to use directly.
This would be a significant change to the buildings sector, but it would dramatically improve our modeling capabilities.
If there are data concerns for the international models, we could include an other category for each appliance type, which could hold the required data that would simplify the model's outputs to represent the availability of data for that country.
Another benefit of doing this would be that we could more easily track HFC consumption, were we to integrate this with the industry sector more completely.