Curtailment Output

Overview

File: curtailment.csv

curtailment.csv records the curtailed generation for every VRE (Variable Renewable Energy) asset at every representative time step. Curtailment is the difference between the maximum available generation (capacity × availability factor) and the actual generation (flow). It represents renewable energy that could have been produced but was not dispatched.

Only assets of type VRE produce rows in this file. If the system has no VRE assets, the file will not be written.

Curtailment is a derived quantity

Curtailment is not an optimization variable — it is computed from the optimal capacity and flow values after the solve. A value of 0.0 means the VRE asset was fully utilized at that time step.

Columns

Column	Type	Description
`commodity`	String	Commodity type generated by the VRE asset (typically `Electricity`)
`zone`	String	Zone (location) where the VRE asset is installed
`resource_id`	String	Unique identifier of the parent VRE asset
`component_id`	String	Unique identifier of the VRE edge component
`resource_type`	String	Asset type (always `VRE` for this output)
`component_type`	String	Type of the edge (e.g., `UnidirectionalEdge{Electricity}`)
`variable`	String	Always `"curtailment"`
`time`	Int	Representative time step index (1-based integer, matches `time` in other output files)
`value`	Float64	Curtailed power at this time step, in the same units as capacity (default: MW)

Calculation

Curtailment at time step $t$ is calculated as:

\[\text{curtailment}(t) = \max\!\Big(0,\; \text{capacity} \times \text{availability}(t) - \text{flow}(t)\Big)\]

where:

capacity is the optimal total installed capacity of the VRE edge (MW), from capacity.csv
availability(t) is the capacity factor at time step $t$ (a value between 0 and 1), from the VRE asset's availability time series input
flow(t) is the actual generation dispatched at time step $t$, from flows.csv

The max(0, ...) ensures curtailment is always non-negative. In practice, the optimizer will only curtail when it is optimal to do so (e.g., when electricity supply exceeds demand and storage is full).

Annual curtailment energy

To compute total annual curtailed energy for an asset, multiply value(t) × weight(t) and sum over all time steps, where weight(t) comes from time_weights.csv:

Annual curtailment (MWh) = Σ_t  curtailment(t) × weight(t) × hours_per_timestep

Configuration

Setting	File	Default	Effect
`OutputLayout` (or `OutputLayout.Curtailment`)	`macro_settings.json`	`"long"`	Set to `"wide"` to pivot time steps into columns.
`WriteFullTimeseries`	`case_settings.json`	`false`	When `true` and TDR is active, also write full-year curtailment to `full_time_series/curtailment.csv`.

Assumptions

VRE assets only. Only assets of type VRE are included. Other asset types (e.g., ThermalPower, Battery) do not produce curtailment rows.
Requires has_capacity = true (true by default). Curtailment is only meaningful when the VRE edge has capacity variables enabled. If the user sets has_capacity = false, the availability × capacity term is zero and curtailment will be 0.0.
Availability time series. The availability (capacity factor) profile is typically specified in the VRE asset's input data as a time series of values in [0, 1] for each representative time step. Note that if no availability time series is provided, Macro defaults to full availability (1.0) at all time steps.
File not written if empty. If the system has no VRE assets, or all VRE edges have has_capacity = false, the curtailment.csv file will not be written. Note: if VRE assets exist but the optimizer assigns them zero capacity, the file is still written (with all-zero values).

Examples

Default Long Format (example rows)

commodity	zone	resource_id	component_id	resource_type	component_type	variable	time	value
Electricity	NE	NE_offshorewind_1	NE_offshorewind_1_edge	VRE	UnidirectionalEdge{Electricity}	curtailment	1	0.0
Electricity	NE	NE_offshorewind_1	NE_offshorewind_1_edge	VRE	UnidirectionalEdge{Electricity}	curtailment	2	45.2
Electricity	NE	NE_offshorewind_1	NE_offshorewind_1_edge	VRE	UnidirectionalEdge{Electricity}	curtailment	3	0.0

Computing Annual Curtailment

using CSV, DataFrames

curtailment = CSV.read("results/curtailment.csv", DataFrame)
weights = CSV.read("results/time_weights.csv", DataFrame)

# Join and compute annual curtailment per asset
df = leftjoin(curtailment, weights, on=:time)
df.annual_MWh = df.value .* df.weight

# Group by asset to get total annual curtailment per VRE asset
annual_by_asset = combine(groupby(df, :resource_id), :annual_MWh => sum => :annual_curtailment_MWh)