feat: add base_grid input to investment cost calculation functions #506
Conversation
| if base_grid is None: | ||
| base_grid = Grid( | ||
| scenario.info["interconnect"].split("_"), source=scenario.info["grid_model"] | ||
| ) |
There was a problem hiding this comment.
if base_grid is not None, do we need to make sure that the grid model is the same than the one in the scenario object?
There was a problem hiding this comment.
@rouille Not sure what you mean by "same". The base grid won't be the same as the one in the scenario object, unless the scenario is run with an empty change table, right?
There was a problem hiding this comment.
I think the purpose of having user defined base_grid here is to provide the capability of comparing the difference between the grid in the scenario and any grid instead of the base_grid in PowerSimdata only.
There was a problem hiding this comment.
@BainanXia, my question is can we have a usa_tamu model in the scenario object, i.e.,scenario.get_grid().grid_model == "usa_tamu"), and a "hifld" model for the grid that is passed, i.e., base_grid.grid_model == "hifld"
There was a problem hiding this comment.
I agree, at the very least we want a check that the grid models are the same. We may also want a check that the interconnections match too, or we need to be sure that we have well-defined behavior for comparing just the overlapping parts of two Grids (e.g. comparing a Western expanded scenario to a USA base scenario).
There was a problem hiding this comment.
Done. We now check for the matching grid model, and we should be able to handle interconnections via the reindexing logic that we add in #507.
| if base_grid is None: | ||
| base_grid = Grid( | ||
| scenario.info["interconnect"].split("_"), source=scenario.info["grid_model"] | ||
| ) |
| if base_grid is None: | ||
| base_grid = Grid( | ||
| scenario.info["interconnect"].split("_"), source=scenario.info["grid_model"] | ||
| ) |
danielolsen
force-pushed
the
daniel/investment_cost_base_grid
branch
2 times, most recently
from
51760ed to
2dafacc
Compare
|
I guess that once we merge #507 we will be able to quickly check that the scenario object that is passed to the functions in the |
I don't think we want to prohibit a Create state scenario from investment cost analysis, do we? All we need is |
danielolsen
force-pushed
the
daniel/investment_cost_base_grid
branch
from
2dafacc to
1ac171f
Compare
Pull Request doc
Purpose
Currently, the user-facing functions to calculate investment cost for each of {ac branches, dc branches, generation/storage} automatically compare against the base Grid in the repository. This PR adds a new parameter
base_gridto each of these functions, so that another Grid can be passed instead.This is useful because the base Grid in the repository does not correspond to any particular point in time, so calculating investment costs between for example a 2021 scenario and a 2030 scenario currently involves calculating the investment costs of each of them compared to the base grid in the repository, and then comparing those differences against each other. If instead the user can say that the 2021 scenario is the 'base' grid to compare against, then all of the implementation details of comparing the two deltas is avoided.
What the code is doing
If a
base_gridis provided to any of the functions, this object is used to calculate investment costs. If this parameter is not specified, then the code proceeds as currently by spinning up a new base Grid.Testing
No tests were performed. Currently, this instantiation of a base Grid limits the testability of the user-facing functions, since we can't pass a MockScenario with a MockGrid. This refactor improves this testability by enabling tests which avoid instantiating a Grid.
Time estimate
10 minutes.