query_master: Query data and aggregate data

Description

This collection of functions retrieves data from the processed PLEXOS solutions and returns it in a convenient format.

Usage

query_master(db, time, col, prop, columns = "name", time.range = NULL, filter = NULL, phase = 4)
query_interval(db, ...)
query_day(db, ...)
query_week(db, ...)
query_month(db, ...)
query_year(db, ...)
sum_master(db, time, col, prop, columns = "name", time.range = NULL, filter = NULL, phase = 4, multiply.time = FALSE)
sum_interval(db, ...)
sum_day(db, ...)
sum_week(db, ...)
sum_month(db, ...)
sum_year(db, ...)

Arguments

PLEXOS database object

time

character. Table to query from (interval, day, week, month, year)

col

character. Collection to query

prop

character vector. Property or properties to query

columns

character. Data columns to query or aggregate by (defaults to name)

time.range

POSIXt or character. Range of dates of length 2 (given as date, datetime or character in 'ymdhms' or 'ymd' format). The Plexos data is assumed to be in UTC so providing a POSIXt vector in a different timezone might cause conflicts. Character vectors are also converted to the UTC format, so here there is not issue.

filter

list. Used to filter by data columns (see details)

phase

integer. PLEXOS optimization phase (1-LT, 2-PASA, 3-MT, 4-ST)

...

parameters passed from shortcut functions to master (all except time)

multiply.time

boolean. When summing interval data, provide the value multiplied by interval duration (See details).

Value

A data frame that contains data summarized/aggregated by scenario.

Details

The family query_* returns the raw data in the databases, while sum_* aggregates the data according to columns.

The functions *_day, *_week, *_month and *_year are shortcuts for the corresponding, *_master function.

The following is a list of valid items for columns and filtering. Additionally, time can be specified for summary data (interval data always includes time).

category
property
name (default for columns)
parent (automatically selected when name is selected)
category
region (only meaningful for generators)
zone (only meaningful for generators)
period_type
band
sample
timeslice

If defined, the filter parameter must be a list. The elements must be chracter vectors and need to have a valid column name (see previous bullet points). For example, one could define it as follows:

filter = list(name = c("Generator1", "Generator2"), region = "Region1")

To filter by time use the time.range parameter, instead of adding it as an entry in the filter parameter. For example use c("2015-03-14", "2015-03-15") in your query. Please note that the year/month/date starts at midnight (00:00:00).

If a scenario has multiple databases, the data will be aggregated automatically. If two or more databases within the same scenario have overlapping time periods, the default is to select the data from the last database (execute summary(db) so see the order). To change this behavior set the global option rplexos.tiebreak to first, last, or all to select data from the first database, the last one or keep all of them.

Multiple properties can be queried within a collection. If prop equals the widcard "*", all the properties within a collection are returned.

The parameter multiply.time allows to multiply values by interval duration (in hours) when doing the sum of interval data. This can be used, for example, to obtain total energy (in MWh) from power time series (in MW).

Examples

Run this code

# Process the folder with the solution file provided by rplexos
location <- location_solution_rplexos()
process_folder(location)

# Query data
db <- plexos_open(location)
query_day(db, "Generator", "Generation")
query_day(db, "Region", "*")
query_interval(db, "Generator", "Generation")