plr_decomposition: Decompose Seasonality from Data

In PVplr: Performance Loss Rate Analysis Pipeline

Decompose Seasonality from Data

Description

Decomposes seasonality from a dataframe that has already passed through a PLR Determination test, e.g. plr_xbx_model. This method has the option of creating plot and data files.

Usage

plr_decomposition(
  data,
  freq,
  power_var,
  time_var,
  plot = FALSE,
  plot_file = NULL,
  title = NULL,
  data_file = NULL
)

Arguments

data	a dataframe containing PV data that has undergone a power predictive model, e.g. plr_xbx_model.
freq	the frequency of seasonality. This is typically 4 but depends on the location of the system.
power_var	name of the power variable, e.g. iacp
time_var	name of the time variable, e.g. tvar
plot	boolean indicating if you wish to save a plot.
plot_file	location to save the plot, if the plot param is given TRUE.
title	the title of the plot created if the plot param is given TRUE.
data_file	location to save data. Currently non-functional.

Value

Dataframe containing decomposed time series features

Examples

#' # build var_list
var_list <- plr_build_var_list(time_var = "timestamp",
                               power_var = "power",
                               irrad_var = "g_poa",
                               temp_var = "mod_temp",
                               wind_var = NA)
# Clean Data
test_dfc <- plr_cleaning(test_df, var_list, irrad_thresh = 100,
                         low_power_thresh = 0.01, high_power_cutoff = NA)
# Perform power modeling step
test_xbx_wbw_res <- plr_xbx_model(test_dfc, var_list, by = "week",
                                  data_cutoff = 30, predict_data = NULL)
                                  
test_xbx_wbw_decomp <- plr_decomposition(test_xbx_wbw_res, freq = 4,
                                         power_var = 'power_var', time_var = 'time_var',
                                         plot = FALSE, plot_file = NULL, title = NULL, 
                                         data_file = NULL)

PVplr documentation built on Feb. 16, 2023, 9:56 p.m.