R/climate.R

In pnnl/capratTX: Thermoelectric capacity at Risk Analysis Tool for ERCOT

#' simulate_gcm_inflows_all_dams
#' @details ...
#' @param data_path path to data directory "ERCOT Reservoir Watershed Delineations and Inflow Scenarios"
#' @param period either "baseline" or "future"
#' @params_path path to calibrated reservoir parameters
#' @importFrom purrr map_dfr
#' @importFrom vroom vroom cols
#' @importFrom dplyr mutate group_by summarise
#' @importFrom lubridate year month
#' @export
#'
simulate_gcm_inflows_all_dams <- function(data_path, period, params_path){

  params <- vroom_silent(params_path)

  paste0(data_path,
         "/Reservoir records and attributes/",
         "ERCOT_reservoir_attributes_and_records_PUBLIC.xlsx") ->
    res_filepath

  read_ERCOT_data(res_filepath)[["storage_levels"]] %>%
    .[["Look Up Name for Reservoir"]] %>%
    unique() -> reservoir_shortnames

  reservoir_shortnames %>%
    .[!grepl("Amistad", .)] %>%
    .[!grepl("STP", .)] %>%
    map_dfr(function(x){

      read_ERCOT_data(res_filepath)[["reservoir_info"]] %>%
        filter(`Look Up Name for Reservoir` == x) %>%
        .[["Maximum storage level (Acre-ft)"]] * af_to_mcm ->
        storage_capacity_reported

      simulate_gcm_inflows(data_path = data_path,
                           reservoir = x,
                           period = period,
                           parameters = params) %>%
        mutate(s_cap = storage_capacity_reported)

    })
}

#' simulate_gcm_inflows
#' @details ...
#' @param data_path path to data directory "ERCOT Reservoir Watershed Delineations and Inflow Scenarios"
#' @param reservoir name of the reservoir
#' @param period either "baseline" or "future"
#' @param parameters tibble with headers of "target", "hedge", and "reservoir"
#' @param output return either optimized parameters ("parameters") or simulation results ("results")
#' @importFrom purrr map_dfr
#' @importFrom dplyr mutate group_by summarise
#' @importFrom lubridate year month
#' @export
#'
simulate_gcm_inflows <- function(data_path, reservoir, period, parameters, s_initial = 1.0){

  paste0(data_path,
         "/Reservoir records and attributes/",
         "ERCOT_reservoir_attributes_and_records_PUBLIC.xlsx") ->
    res_filepath

  parameters %>% filter(reservoir == !! reservoir) -> x
  target <- x$target[1]
  hedge <- x$hedge[1]

  # get storage capacity
  read_ERCOT_data(res_filepath)[["reservoir_info"]] %>%
    filter(`Look Up Name for Reservoir` == reservoir) %>%
    .[["Maximum storage level (Acre-ft)"]] * af_to_mcm ->
    storage_capacity

  gcm_names %>%
    map_dfr(function(gcm){

      # read the inflows
      read_gcm_flows(reservoir, gcm, period, data_path) %>%
        mutate(year = year(date), month = month(date)) %>%
        group_by(year, month) %>%
        summarise(cumulative_inflow_mcm = sum(flow)) %>% ungroup() ->
        flows

      # simulate
      simulate_reservoir(
        s_cap = storage_capacity,
        s = storage_capacity * s_initial,
        i = flows[["cumulative_inflow_mcm"]],
        params = c(target, hedge)) %>%
        select(-s_obs) %>%
        mutate(reservoir = !!reservoir,
               gcm = !!gcm,
               inflow = flows[["cumulative_inflow_mcm"]],
               year = flows[["year"]],
               month = flows[["month"]])
    })
}

#' generate_monthly_flow_replicates
#' @details ...
#' @param reservoir name of the reservoir
#' @param data_path path to data directory "ERCOT Reservoir Watershed Delineations and Inflow Scenarios"
#' @importFrom purrr map_dfr
#' @importFrom reservoir dirtyreps
#' @import dplyr
#' @importFrom lubridate year month
#' @export
#'
generate_monthly_flow_replicates <- function(reservoir, data_path){

  read_controlflows(reservoir, data_path) %>%
    mutate(year = year(date), month = month(date)) %>%
    group_by(year, month) %>%
    summarise(obs_inflow = sum(flow)) %>% ungroup() %>%
    arrange(year, month) -> monthy_inflow

  monthy_inflow %>% .[["obs_inflow"]] %>%
    ts(start = monthy_inflow[["year"]][1], frequency = 12) ->
    monthly_inflow_ts

  dirtyreps(monthly_inflow_ts, n_reps) -> replicates

  bind_cols(monthy_inflow, as_tibble(replicates)) %>%
    mutate(date = as_date(paste(year, month, 1, sep = "-"))) %>%
    select(-year, -month) %>%
    gather(rep, flow, -date) ->
    reps

  reps %>% ggplot(aes(date, flow, col = rep)) +
    geom_line(alpha = 0.5) +
    scale_color_manual(values = c("blue", rep("grey", n_reps))) +
    theme_void()

    ymd("2001-01", truncated = 1)
}