Skip to contents

Compare solved scenarios (or runs of one scenario)

Source: R/compare_scenarios.R
compare_scenarios.Rd

Builds a results comparison across a named list of solved scenarios, or across recorded runs of a single scenario (see scenario_runs()). One element is the base (default: the first); every other element is compared against it: objectives and run metadata in an overview table, and selected solved variables value-by-value through a tolerance-aware full join at an annual aggregation level.

Usage

compare_scenarios(
  object,
  runs = NULL,
  name = NULL,
  base = 1L,
  level = c("annual_region", "annual_total"),
  tol_rel = 1e-06,
  tol_abs = 1e-06,
  inputs = FALSE,
  top_n = 10L
)

# S3 method for class 'scenarios_cmp'
print(x, top = 10L, ...)

# S3 method for class 'scenarios_cmp'
autoplot(
  object,
  type = c("objective", "generation", "capacity", "new_capacity", "fuel", "emissions",
    "cost", "delta"),
  comm = "ELC",
  region = NULL,
  year = NULL,
  position = c("facet", "dodge"),
  variable = NULL,
  top_n = 12,
  ...
)

# S4 method for class 'scenarios_cmp'
plot(x, y, ...)

Arguments

object

a scenarios_cmp object.

runs

character, run ids of scen to compare (as listed by scenario_runs()); each is read into a local copy via read_solution() – the caller's scenario keeps its active run.

name

character, variables to compare (overrides the role-driven default).

base

integer position or element name of the baseline (default: the first element).

level

aggregation level for value comparisons: "annual_region" (sum over timeslices) or "annual_total" (also over regions; variables with role "interregional" are skipped there).

tol_rel, tol_abs

numeric comparison tolerances (a value differs when abs(diff) > tol_abs + tol_rel * max(abs(values))).

inputs

logical, also diff the model inputs of every scenario against the base via compare_inputs().

top_n

integer, passed to getMix() for the mix charts: keep the top_n largest processes per scenario and lump the rest into "Other". NULL/Inf keeps all.

x

a scenarios_cmp object (plot delegates to autoplot).

top

integer, number of top differences to print.

...

ignored.

type

chart type: "objective" (bars per scenario), one of the getMix() types ("generation", "capacity", "new_capacity", "fuel"), "emissions", "cost" (role-driven cost breakdown per scenario), or "delta" (largest value differences).

comm, region, year

filters forwarded to getMix().

position

"facet" (stacked mixes, one panel per scenario) or "dodge" (scenarios side by side, aggregated over process).

variable

for type = "delta": restrict to one variable.

y

ignored.

Value

A scenarios_cmp object (a list) with components: overview (one row per scenario: run/solver/status/objective and objective differences vs base), values (per variable x scenario: n_diff, max_abs_diff, max_rel_diff, status), deltas (named list per variable of row-level differences), top (the top_n largest absolute differences), inputs (list of compare_inputs() results when inputs = TRUE), and the carried scen list / base / level / tol. The object holds the full scenario list so autoplot() and report() can rebuild mixes from it. print() gives a compact report; autoplot() draws comparison charts; report() renders the comparative report.

Details

Variables default to the role-driven selection the scenario report uses: every solved variable the catalogue declares as a cost (minus vObjective / vTotalCost) plus the key quantity variables (vTechCap, vTechNewCap, vTechOut, vEmsFuelTot) that are present.

See also

compare_inputs(), compare_models(), getMix() (which also accepts named scenario lists), scenario_runs().

Other comparison: compare_inputs(), compare_models()

Examples

if (FALSE) { # \dontrun{
open_project(tempdir())

ek  <- topia$modules$electricity$R3
mod <- newModel("TOPIA", data = ek$repo,
                calendar = topia$modules$calendars$s4_h24,
                region = ek$regions,
                horizon = topia$modules$horizons$base,
                discount = 0.05)
mod <- setGeoscale(mod, ek$geoscale)

base <- solve_model(mod, name = "BASE", solver = solver_options$glpk)
cap  <- solve_model(add(mod, ek$CO2_CAP), name = "CAP",
                    solver = solver_options$glpk)

# a NAMED list, the first entry as the baseline
cmp <- compare_scenarios(list(BASE = base, CAP = cap))
cmp
ggplot2::autoplot(cmp, type = "generation", comm = "ELC")

# or several runs of ONE scenario folder, by run id
compare_scenarios(base, runs = scenario_runs(base)$run)
} # }