PyPSA
State-level power sector model
Feature Request
Currently PyPSA-USA only provides county-level and ReEDS BA-level power sector modeling. It would be so cool if it can model the power sector at the state-level.
Suggested Solution
The current cluster_network.py only supports reeds_zone and county for topological_boundaries. Adding state support requires itl_state and itl_costs_state files, but the code link used to generate the data (https://github.nrel.gov/pbrown/TSC) is broken.
Maybe we can generate state-level transmission data by aggregating existing reeds_zone data, and this would involve:
- Sum capacities across all zone pairs between states
- Distances calculated based on geometric centers and unit costs calculated based on distance-weighed averaging
However, I'm uncertain whether this aggregation approach would introduce significant errors, and whether there are better alternatives. Besides, I'm not sure if further changes are needed.
Additional Info
No response
I like the idea! This would be fairly straightforward to implement in cluster_network by defining a new custom_busmap which aggregates all the state level buses together! I can provide more detail or we can chat about this at the next call if you would like.
@trevorb1 @ktehranchi I just noticed that in rule cluster_network, itl_costs_county uses AC but itl_costs_reeds_zone uses DC. What's the consideration here? Is it that for longger transmissions, DC is cheaper than AC? Aren't we assuming the carrier of the majority of the transmission lines to be AC, tough? I'm concerned with this because we can see from the distance_cost file that DC is much cheaper than AC even at ~100 miles length. Usually DC is economicaly compariable to AC only when the length is over 600-800 km. So I suspect that the DC cost here doesn't include the cost of HVDC inverters, and we should use AC distance_cost data instead.
transmission_distance_cost_500kVdc_ba.csv
transmission_distance_cost_500kVac_ba.csv