E4ST Inputs

[Ethan-Russell]

How we currently handle inputs:

• ESC file - E4ST Case file - tells us how to prepare MPC file: new generators, policies, future prices, etc.
• MPC file - information of the system: existing generators, buses, etc.
• Occasionally other tables not in the ESC file, such as for carbon storage steps

Starting with a list of all of the input tables which, required or optional, can go into E4ST, using their current names

In MPC file

Note: the names of tables are suffixed with sub because they are based on the set of substations, after having buses at the same substations combined.

• mpc_vars gives instructions for how to read sheets of MPC file into mpc structure in matlab. Also gives descriptions of columns in those sheets
• mpc Only contains baseMVA single scalar value. Probably separate due to matpower mpc format
• bus_sub Table of buses and their electrical attributes. Contains lat and lon, but that info is also in sub_map
• branch_sub Table of branches (AC transmission lines) and their attributes
• dcline_sub Table of DC transmission lines and their attributes
• gens_sub Table of existing generators, including load modeled as generators (dl). Includes emissions rates, costs, and generator-specific location information.
• sub_map Locational information for substations. As a separate sheet from bus_sub because locational information is used to join onto new generators, as new generators get information of the bus they are connected to.

💡 Currently, the dl generators tell us how load is distributed among buses. In the future, we can use this information directly without having to model load as generators. 💡 In development, should keep in mind how we can do endogenous transmission expansion, as this might require some additional input information

In ESC File

• esc_vars Analogous to mpc_vars
• esc contains several vectors and constants, which really could be options
• custom_map A sheet that defines new "custom" areas to add to the bus_map or sub_map, by combining and building on other existing area columns
• hrs_map Table of representative hours, their attributes, and groupings into representative days ⚠️ many inputs to E4ST depend on the set of representative hours
• cobra_map Contains by-county results of a linearization of COBRA, used to get quick health damages on generator results 💡 We can run COBRA separately, and also will incorporate INMAP, so not so important.
• state_fixed A table of some state-based data. Added relatively recently for project-specific needs
• state_year A table of state-based data which also depends on year. Added relatively recently for project-specific needs
• interface_lim Defines interface lims, as limit on sum of AC transmission going directly between two areas. These typically are actual details of how the grid is operated, so belongs with grid information.
• gens_build_sub Lists buildable generators, their costs and emissions. Can be sets or individual generators. ⚠️ right now the cost values must be in alignment with the annual adjustments made in gen_ann_adj
• gen_tgt_af Lists targeted average availability factor for sets of generators. Average over both generators and hours.
• gen_ann_adj Does lots of things, most of which involve values which change depending on years. It prevents some sets of generators from retiring in certain years. It sets some values, both generator attributes and constants, which vary by year (and some which don't). It applies "routine_capex" and "life extension costs". It alters the capital and FOM costs over time of newly built generators. It applies regional multipliers to the capital cost of new generators (doesn't vary with time).
• gen_af Lists hourly availability factors for sets of generators. Note, the gen_af and gen_tgt_af sheets both get used to create an availability factor matrix which has the value for each generator in each hour. Also, note that the wind and solar values are no longer set in this sheet but in the update_rephrs code.
• gen_cf Sets min and max average capacity factor for sets of generators. Only used for fuel-limited resources where we would want to limit capacity factors
• gen_pmin sets PMIN for sets of generators. Currently all values are zero, so not sure when this would be useful.
• fuel_cost Sets fuel costs by year, area, and fuel type
• cap_match Sets some constraints on the total active capacity of some sets of generators. Should be re-thought, as capacity requirements should be listed with policies, and forced retirements shouldn't be managed with constraints.
• policy Lists policies of various types to add to the simulation. ⚠️ Different policies need different levels of numerical input, so the values in this sheet might mean different things for different policies. How do we want to deal with this?
• load_growth Table of annual growth rate in electricity load in regions. Usefulness is questionable when we use load_match with finer detail, but potentially could still add a dimension of realism.
• load_shape Information for how load is spread among representative hours, by regions
• load_match Sets what actual total annual load should be in a year, by regions
• load_add Used for adding load, after all adjustments from load growth, shape, and match.

Elsewhere

• CO2 Storage Steps For each state, list of best CO2 storage options, with total capacity, storage cost, and transportation cost. Read in RunSim_UserAux
• Biomass supply and transport distances Not used recently, but i believe parameters for modeling cofiring of woody biomass at existing coal plants. Read in RunSim_UserAux
• Fossil Fuel VOM calibrations For the calibration that we do of natural gas and coal operation depending on regulation and market status. Read in RunSim_UserAux
• Voluntary green power Estimates of green power purchased voluntarily, on top of state policies. Read in user_pol_prjspecifc
• distribution costs While this was never used, was an attempt once to have different elec distribution costs in different areas. Read into result_bus

The above list does not include various numerical assumptions that may be in the model but not in a table, like the CCS retrofit cost curves.

How others handle inputs:

• GenX - multiple CSV files, as in here
• REISE
• ReEDS - multiple CSV files, a cases.csv file where you can change most features that would differ between scenarios (saved with a new name and then directed to at the start of the run)

How we could handle inputs:

Goals:

• Be aware of what goes into each simulation! (currently we copy the inputs into the output folder)

[Ethan-Russell]

Things that might get changed for a given project:

• policy
• load_match

Things that get changed less frequently:

• representative hours
• mpc file (only updates when we change generators, or the grid, etc)

[Ethan-Russell]

What is the set of information we might need to run an instance of E4ST?

Top level config file:

GenFile: L:/...
LoadFile: L:/...
Extras:
    [
        AvailabilityFactor:
            Filename: L:/whatever
        CESPolicy:
            bm: 0.5
            bm_unit: kg/MWh

    ]

Say you want to run a bunch of cases for a specific project. In code, you could do something like:

config = loadconfig(baseconfigfile)
for load_profile in load_profiles
   config["LoadFile"] = load_profile
   runE4st(config)
end

[Ethan-Russell]

Closing for now - see e4st-input-processing repo