New-Build Resources Settings¶

These parameters control technology costs, new resource definitions, regional availability, and capacity constraints for candidate generators.

Technology Cost Sources¶

`new_resources`¶

Type: List of lists Required: For ATB-based technologies Example: See below

List of technologies from NREL Annual Technology Baseline (ATB) to include as new-build candidates. Each item is a list: [technology, tech_detail, cost_case, size_mw].

new_resources:
  - [NaturalGas, CCCCSAvgCF, Moderate, 500]
  - [NaturalGas, CTAvgCF, Moderate, 100]
  - [Battery, "*", Moderate, 100]
  - [UtilityPV, Class1, Moderate, 100]
  - [LandbasedWind, Class3, Moderate, 100]
  - [OffshoreWind, Class3, Moderate, 100]
  - [Nuclear, Nuclear - Large, Moderate, 1000]

Format: [technology, tech_detail, cost_case, size_mw]

technology: Technology type (e.g., NaturalGas, Battery, UtilityPV)
tech_detail: Specific variant (e.g., CCAvgCF, Class1, * for wildcard)
cost_case: Cost trajectory (Conservative, Moderate, Advanced)
size_mw: Unit size in MW

`resource_data_year`year`¶

Type: Integer Required: No Default: Latest available Example: 2023

Which year's ATB data to use for technology costs.

resource_data_year: 2023

ATB is updated annually with revised cost projections. Specify the data vintage for reproducibility.

`resource_financial_case`case`¶

Type: String Required: No Default: "Market" Example: "Market"

ATB financial assumptions case. Options:

Market: Market-based financing
R&D: Optimistic R&D case with policy support

resource_financial_case: Market

`resource_cap_recovery_years`¶

Type: Integer Required: No Default: 20 Example: 20

Default capital recovery period (economic lifetime) for new generators, used to calculate annualized capital costs.

resource_cap_recovery_years: 20

`alt_resource_cap_recovery_years`¶

Type: Dictionary (technology → years) Required: No Example: See below

Override capital recovery period for specific technologies (matched using string contains).

alt_resource_cap_recovery_years:
  NaturalGas_CCAvg: 15
  NaturalGas_CTAvg: 15
  Battery: 15
  Nuclear: 40

Cost and Performance Adjustments¶

`resource_modifiers`¶

Type: Dictionary (tech_name → modifications) Required: No Example: See below

Modify ATB technology parameters. Each entry specifies technology/tech_detail matching and parameter changes.

resource_modifiers:
  batteries:
    technology: Battery
    tech_detail: "*"
    Var_OM_Cost_per_MWh: [add, 0.15]
    Var_OM_Cost_per_MWh_In: 0.15
    size_mw: 100

Each entry creates a candidate resource with the specified unit size.

Parameter values inside each modifier can also be specified as year-keyed dictionaries (see Settings by Model Year) to vary costs across planning periods without a full scenario management setup:

resource_modifiers:
  batteries:
    technology: Battery
    tech_detail: "*"
    Var_OM_Cost_per_MWh:
      2030: [add, 0.15]
      2040: [add, 0.10]
    size_mw: 100

`modified_new_resources`¶

Type: Dictionary (new_name → base + modifications) Required: No Example: See below

Create new technology variants by copying and modifying existing ATB technologies.

modified_new_resources:
  NGCCS100:
    technology: NaturalGas
    tech_detail: CCCCSAvgCF
    cost_case: Conservative
    size_mw: 500
    new_technology: NaturalGas
    new_tech_detail: CCS100
    new_cost_case: Moderate
    capex_mw: [add, 116000]
    heat_rate: [add, 0.365]
    fixed_o_m_mw: [add, 9670]
    variable_o_m_mwh: [mul, 1.076]

  hydrogen_turbine:
    technology: NaturalGas
    tech_detail: 1-on-1 Combined Cycle (H-Frame)
    cost_case: Moderate
    size_mw: 100
    new_technology: hydrogen
    new_tech_detail: 1-on-1 Combined Cycle (H-Frame)
    new_cost_case: Moderate

Base parameters (copy from existing ATB technology):

technology: ATB technology to copy from
tech_detail: ATB tech detail to copy from
cost_case: ATB cost case to copy from
size_mw: Unit size

New parameters (create new technology with this name):

new_technology: New technology name
new_tech_detail: New tech detail name
new_cost_case: New cost case name

Modification operators:

[add, value]: Add to base value
[mul, value]: Multiply base value
[sub, value]: Subtract from base value
[truediv, value]: Divide base value
Direct value: Override base value

Settings by Model Year¶

Year-keyed values are supported across all settings sections, not just new-build parameters.

Use Year-Keyed Values for the complete rules, validation behavior, and additional examples.

Quick summary:

Exact year values take priority.
default is a fallback key.
Incomplete year coverage without fallback raises a ValueError.

Example in resource_modifiers:

resource_modifiers:
  utility_pv:
    technology: UtilityPV
    tech_detail: Class1
    capex_mw:
      2030: [mul, 0.9]
      2040: [mul, 0.8]

Example with fallback:

carbon_tax:
  2030: 25
  default: 50

Note

The settings_management key itself is never modified by year-keyed resolution; it is always processed by scenario management logic first.

Regional Availability¶

`new_gen_not_available`¶

Type: Dictionary (region → list of technologies) Required: No Example: See below

Exclude specific new-build technologies from certain regions.

new_gen_not_available:
  AZ:
    - OffshoreWind_Class1_Moderate_-1
    - Geothermal_HydroFlash_Moderate
  CA_N:
    - Coal_new_Moderate

Use cases:

Physical constraints (no offshore wind in landlocked states)
Policy constraints (coal banned)
Resource availability (no geothermal potential)

Interconnection Costs¶

`interconnect_capex_mw`¶

Type: Number or dictionary Required: No Default: None (falls back to legacy transmission_investment_cost if provided) Example: See patterns below

Flexible specification of per-MW interconnection capital cost (USD/MW) applied to new and existing resources. Replaces the deprecated spur line mileage system (transmission_investment_cost with capacity_limit_spur_fn file).

Important: All cost values must be in the same dollar year as target_usd_year (the target dollar year for cost normalization). PowerGenome does not automatically adjust interconnection costs for inflation.

Supported Patterns:

1. Scalar¶

Apply uniform cost to all resources:

interconnect_capex_mw: 150000  # $150k/MW for all resources

2. Explicit Schema (Recommended)¶

Use an explicit fallback with typed override blocks:

interconnect_capex_mw:
  fallback_capex_mw: 110000
  by_technology:
    wind: 120000
    offshore_wind: 200000
    battery: 50000
  by_region:
    CA_N: 125000
  by_region_technology:
    CA_N:
      offshore_wind: 210000

Precedence (most specific to least specific):

by_region_technology
by_region
by_technology
fallback_capex_mw

3. Region-Only¶

Different costs by region (exact region name matching):

interconnect_capex_mw:
  fallback_capex_mw: 120000
  by_region:
    CA_N: 140000
    CA_S: 130000
    AZ: 125000

4. Technology-Only¶

Different costs by technology using case-insensitive substring matching with shortest-first precedence:

interconnect_capex_mw:
  fallback_capex_mw: 100000
  by_technology:
    wind: 120000          # matches 'LandbasedWind_Class3_Moderate', 'OffshoreWind_Class1_Moderate'
    offshore_wind: 200000 # longer substring overwrites previous 'wind' assignment
    battery: 50000        # matches 'Battery_4Hr_Moderate', 'Battery_*_Moderate'
    solar: 110000         # matches 'UtilityPV_Class1_Moderate'

Precedence rule: Shorter substrings applied first, then longer (more specific) substrings override previous assignments. This allows general categories with specific exceptions.

5. Region -> Technology Nested¶

Region-specific technology costs with regions as keys under by_region_technology:

interconnect_capex_mw:
  fallback_capex_mw: 110000
  by_region_technology:
    CA_N:
      battery: 60000
      offshore_wind: 210000
      solar: 105000
    CA_S:
      solar: 95000
      wind: 115000
  by_region:
    AZ: 125000

Within each region, technology substrings follow shortest-first precedence.

Matching Rules:

Technology matching: Case-insensitive substring search against the technology column
Substring precedence: Shortest substrings processed first; longer substrings override
Region matching: Exact match against model region names (case-sensitive)
Explicit schema: Use only fallback_capex_mw, by_region, by_technology, and by_region_technology at the top level.
Invalid mixing: In legacy syntax, mixing region and technology keys at the same top level is invalid.

Examples of Invalid Configuration:

# ERROR: Cannot mix region and technology keys
interconnect_capex_mw:
  CA_N: 140000      # Region key
  wind: 120000      # Technology key - mixing not allowed!

Legacy Syntax (Deprecated):

You may still use the older top-level default style for backward compatibility. A deprecation warning is logged at runtime.

Behavior:

Non-destructive: Existing non-zero interconnect_capex_mw values in the resource dataframe are never overwritten
Automatic annuity: interconnect_annuity is computed automatically for:
Newly assigned capex rows
Pre-existing capex rows with zero/blank annuity
Financial parameters: Annuity calculation uses plant-specific wacc_real and cap_recovery_years from the resource dataframe
Legacy bypass: When interconnect_capex_mw is provided, legacy spur mileage columns (spur_miles, offshore_spur_miles, tx_miles) are ignored (warning logged if present)

Migration from Legacy System:

Old approach (deprecated):

# Settings
transmission_investment_cost:
  spur:
    capex_mw_mile:
      CA_N: 3000
      CA_S: 3200
    wacc: 0.069
    investment_years: 60

# Required extra_inputs file: capacity_limit_spur_fn
# File: resource_capacity_spur.csv
# resource,region,spur_miles
# offshore_wind_fixed,CA_N,45
# solar_pv,CA_N,12

New approach (recommended):

interconnect_capex_mw:
  fallback_capex_mw: 120000
  by_technology:
    offshore_wind: 200000  # Direct cost specification
    solar: 105000

Benefits of new system:

No external file dependency
More intuitive (direct $/MW vs. mileage calculation)
Flexible pattern matching (region, technology, or nested)
Automatic annuity handling
Better error messages

`transmission_investment_cost`¶

Type: Dictionary Required: No Deprecated: Use interconnect_capex_mw instead Example: See legacy documentation

Legacy spur line mileage-based interconnection cost system. Contains nested dictionaries for spur, offshore_spur, and tx with keys:

capex_mw_mile: Cost per MW-mile (by region)
wacc: Weighted average cost of capital
investment_years: Capital recovery period

Deprecation notice: This parameter is deprecated and will be removed in a future release. It is only used when interconnect_capex_mw is not provided, and emits warnings during execution. Migrate to interconnect_capex_mw for simplified, more flexible cost specification.

Renewable Resource Clustering¶

`renewables_clusters`¶

Type: List of dictionaries Required: For renewable resources (wind, solar, geothermal) with spatially-distributed potential Example: See below

Configure clustering of renewable resource sites from external datasets (e.g., NREL reV). PowerGenome filters, bins, and clusters pre-existing resource data to create representative sites for the model.

renewables_clusters:
  - region: all
    technology: landbasedwind
    filter:
      - feature: lcoe_interconnect_adj
        max: 60
    bin:
      - feature: lcoe_interconnect_adj
        weights: capacity_mw
        mw_per_bin: 10000
    cluster:
      - feature: cf
        n_clusters: 2
        method: agg

  - region: p4
    technology: utilitypv
    filter:
      - feature: lcoe_interconnect_adj
        max: 35
    bin:
      - feature: lcoe_interconnect_adj
        weights: capacity_mw
        mw_per_bin: 50000
    cluster:
      - feature: lcoe_interconnect_adj
        n_clusters: 2
        method: agg

  - region: all
    technology: geothermal
    type: egs
    filter:
      - feature: class
        max: 6
    group:
      - class
    group_modifiers:
      - group: class
        group_value: 6
        Inv_Cost_per_MWyr: [add, 64820]

  - region: all
    technology: offshorewind
    turbine_type: floating
    pref_site: 0
    bin:
      - feature: lcoe_interconnect_adj
        weights: capacity_mw
        q: 2

Common parameters:

region: Model region name or all (applies to all regions)
technology: Technology type (lowercase: landbasedwind, utilitypv, offshorewind, geothermal)

Filtering (filter):

feature: Attribute to filter on (lcoe, cf, lcoe_interconnect_adj, class)
max: Maximum value threshold (exclude resources above this)
min: Minimum value threshold (exclude resources below this)

Binning (bin):

Groups resources into bins before clustering:

feature: Attribute to bin by (lcoe, lcoe_interconnect_adj, cf)
weights: Weight bins by capacity (mw, capacity_mw)
q: Number of quantile bins (equal-frequency binning)
mw_per_bin: Fixed MW capacity per bin (alternative to q)
bins: Custom bin edges (list of values)

Clustering (cluster):

Aggregates resources within each bin:

feature: Attribute to cluster by (cf, lcoe, lcoe_interconnect_adj, profile)
n_clusters: Number of clusters to create
method: Clustering algorithm (agg for agglomerative hierarchical clustering)

Grouping (group + group_modifiers):

Alternative to binning/clustering for technologies with discrete classes:

group: List of features to group by (e.g., [class])
group_modifiers: List of cost adjustments per group value

Technology-specific parameters:

Offshore wind: turbine_type (fixed, floating), pref_site (0 or 1 for preferred sites)
Geothermal: type (egs, geohydrobinary)

Workflow:

Load resource sites from external data (NREL reV, ReEDS)
Filter sites by quality thresholds
Bin remaining sites by cost/performance
Cluster sites within each bin to reduce model size
Create aggregated resource with weighted-average characteristics

See Configure Renewable Clusters How-To for detailed examples.

Cost Multipliers¶

Regional capital cost multipliers adjust technology costs by region to account for differences in labor costs, permitting, and construction conditions. PowerGenome uses the regional_cost_factor_table to apply these adjustments.

See Regional Configuration - Cost Multipliers for details on cost_multiplier_region_map and cost_multiplier_technology_map parameters.

Example Configuration¶

Complete new-build configuration:

# New-build technologies
resource_data_year: 2023
resource_financial_case: Market
resource_cap_recovery_years: 20

new_resources:
  - [NaturalGas, CCCCSAvgCF, Conservative, 500]
  - [NaturalGas, CTAvgCF, Moderate, 100]
  - [Battery, "*", Moderate, 100]
  - [UtilityPV, Class1, Moderate, 100]
  - [LandbasedWind, Class3, Moderate, 100]
  - [Nuclear, Nuclear - Large, Moderate, 1000]

# Technology modifications
resource_modifiers:
  batteries:
    technology: Battery
    tech_detail: "*"
    Var_OM_Cost_per_MWh: [add, 0.15]

# Regional restrictions
new_gen_not_available:
  AZ:
    - OffshoreWind_Class1_Moderate
  CA_N:
    - Coal_newAvgCF_Moderate

# Capacity limits
new_build_max_capacity:
  CA_N:
    Nuclear_Nuclear_Moderate: 2000
    UtilityPV_Class1_Moderate: 10000

# Renewable sites
renewable_clusters:
  - region: CA_N
    technology: LandbasedWind
    cluster: 1
    max_capacity: 2000
    profile_id: CA_N_wind_class3

RESOURCE_GROUP_PROFILES: /data/nrel_profiles

Model Definition: model_year affects cost projections
Regions: Regional availability and cost multipliers
Fuels: Fuel costs for thermal technologies
Resource Tags: Tag assignment for new resources

New-Build Resources Settings¶

Technology Cost Sources¶

new_resources¶

resource_data_yearyear`¶

resource_financial_casecase`¶

resource_cap_recovery_years¶

alt_resource_cap_recovery_years¶