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

capratTX

The goal of capratTX is to simulate water storage reservoirs providing cooling water to thermal power plants in the ERCOT.

Please use issues for Q & A.
For other information please contact Sean Turner (sean.turner@pnnl.gov)

You must download the ERCOT Reservoir Watershed Delineations and Inflow Scenarios data files to run this software.

You can install the development version of capratTX using devtools:

library(devtools)
install_github("pnnl/capratTX")

Start by loading the library and setting the data path:

library(capratTX)
data_path <- "your_directory/ERCOT Reservoir Watershed Delineations and Inflow Scenarios/"

data_path <- "../../WECC-ERCOT/ERCOT/data/ERCOT Reservoir Watershed Delineations and Inflow Scenarios/"

Calibrate reservoirs and save a parameter file (calibration may take a minute or two reservoir operating functions are optimized):

calibrate_all_reservoirs(data_path) -> calibrated_parameters
#> Arlington
#> Arrowhead
#> Bardwell
#> Belton
#> Bridgeport
#> Buchanan
#> Canyon
#> Colorado City
#> Eufaula
#> Falcon
#> Graham
#> Granbury
#> Houston
#> Joe Pool
#> Kemp
#> Lavon
#> Limestone
#> Livingston
#> Palo Pinto
#> Pat Mayse
#> Ray Hubbard
#> Ray Roberts
#> Richland-Chambers
#> Texana
#> Texoma
#> Toledo Bend
#> Whitney
#> Coleto Creek
#> Martin
#> Travis
#> Waco
head(calibrated_parameters)
#> # A tibble: 6 x 3
#>   target hedge reservoir 
#>    <dbl> <dbl> <chr>     
#> 1  0.897 1.05  Arlington 
#> 2  0.955 0.447 Arrowhead 
#> 3  0.498 1.05  Bardwell  
#> 4  0.428 0.816 Belton    
#> 5  0.926 0.472 Bridgeport
#> 6  0.901 0.437 Buchanan
readr::write_csv(calibrated_parameters, "params.csv")
#> Registered S3 methods overwritten by 'readr':
#>   method           from 
#>   format.col_spec  vroom
#>   print.col_spec   vroom
#>   print.collector  vroom
#>   print.date_names vroom
#>   print.locale     vroom
#>   str.col_spec     vroom

To simulate all dams with the calibrated rules under a given period (“baseline” or “future”):

simulate_gcm_inflows_all_dams(data_path, period = "future",
                              params_path = "params.csv") -> sim_data
head(sim_data)
#> # A tibble: 6 x 8
#>   s_sim r_sim reservoir gcm    inflow  year month s_cap
#>   <dbl> <dbl> <chr>     <chr>   <dbl> <dbl> <dbl> <dbl>
#> 1  49.6  1.97 Arlington ACCESS   1.97  2011     1  49.6
#> 2  49.6  2.04 Arlington ACCESS   2.04  2011     2  49.6
#> 3  49.6  5.80 Arlington ACCESS   5.80  2011     3  49.6
#> 4  49.6  4.08 Arlington ACCESS   4.08  2011     4  49.6
#> 5  49.6  4.71 Arlington ACCESS   4.71  2011     5  49.6
#> 6  49.6  2.23 Arlington ACCESS   2.23  2011     6  49.6

Plot storage projections for a particular reservoir:

library(ggplot2)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

sim_data %>% 
  # select reservoir desired
  filter(reservoir == "Travis") %>% 
  # get annual minimums
  group_by(reservoir, gcm, year) %>% 
  summarise(s_min = min(s_sim)) %>% 
  ungroup() %>% 
  ggplot(aes(year, s_min, col = gcm)) +
  geom_line() +
  labs(y = "Minimum storage level (MCM)", x = NULL)