This repository contains the analysis and visualization for our water-informed OPF paper. More information about this project can be found here. That url also provides all the pre-computed input and output files needed for this repository (see section II below).
water-OPF
│ .gitignore
│ config.ini: Configuration file needed to run `main.py`
│ LICENSE
│ main.py: Python script to run contents of `analysis.py` and `viz.py`
│ README.md
│ run.sh: Bash script to run `main.py`
│ slurm_nonuniform_run.sh: Slurm batch file to run nonuniform sensitivity analysis
│ slurm_uniform_run.sh: Slurm batch file to run uniform sensitivity analysis
│ water-OPF.yml: Conda environment
│
├───QGIS Manual Spatial Analysis and Cartography
│ Generator Locations.qgz: QGIS project for manual matching synthetic and EIA generators
│
├───src: Source code directory
│ analysis.py: Functions used for analysis
│ to_MATPOWER.m: MATLAB script to convert matpower format
│ viz.py: Functions used to visualization
│
└───water-OPF-io-v2.3
├───figures: Figures that appear in the paper
│ uniform_water_coefficient_distribution.pdf
│ region_water_boxplots.pdf
│ coal_scatter_kmeans.pdf
│ hnwc_histograms.pdf
│ historic_load_hist.pdf
│ effect_of_withdrawal_weight_on_withdrawal.pdf
│ effect_of_withdrawal_weight_plant_output.pdf
│ effect_of_withdrawal_weight_line_flows.pdf
│ decision_tree_total_cost
│ decision_tree_total_cost.svg
│ decision_tree_generator_cost
│ decision_tree_generator_cost.svg
│ decision_tree_withdrawal
│ decision_tree_withdrawal.svg
│ decision_tree_consumption
│ decision_tree_consumption.svg
│ nonuniform_sobol_heatmap.pdf
│
├───manual_files: Files generated via manual processes
│ gen_matches.csv: Generated matching
│ operational_scenarios.csv: Operational scenarios for nonuniform SA
│
├───tables: Tables that appear in the paper
│ line_flows.csv
│ system_information.csv
│
├───generated_files: generated files with 'checkpoint' files useful for debugging
│ │ case.p
│ │ case_match.p
│ │ case_match_water.p
│ │ hnwc.csv
│ │ case_match_water_optimize_ready.p
│ │ nonuniform_sa_results.csv
│ │ nonuniform_sa_sobol.csv
│ │ uniform_sa_results.csv
│ │
│ └───synthetic_grid
│ case.mat
│ gen_info.csv
│
├───figures_manual: Figures manually created.
│ Synthetic Generator Map.pdf
│ decision_tree_total_cost.pdf
│ decision_tree_consumption.pdf
│ decision_tree_generator_cost.pdf
│ decision_tree_withdrawal.pdf
│
└───external_data: External data sources with sources
├───EIA_theremoelectric_water_use
│ cooling_detail_2018.xlsx
│ cooling_detail_2017.xlsx
│ cooling_detail_2016.xlsx
│ cooling_detail_2015.xlsx
│ cooling_detail_2019.xlsx
│ README.txt
│ cooling_detail_2014.xlsx
│
├───load_exogenous_parameter
│ 20180101-20200101 MISO Forecasted Cleared & Actual Load.csv
│ README.txt
│
├───EIA_PowerPlants_Locations
│ PowerPlants_US_202004.prj
│ PowerPlants_US_202004.dbf
│ PowerPlants_US_202004.shx
│ PowerPlants_US_202004.shp
│ PowerPlants_US_202004.cpg
│ PowerPlants_US_202004.sbn
│ PowerPlants_US_202004.sbx
│ PowerPlants_US_202004.shp.xml
│ README.txt
│
├───Illinois Synthetic Grid Gens
│ gens.shp
│ gens.dbf
│ gens.shx
│ Source.txt
│
├───North American Rivers and Lakes Illinois 2019 Flows
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.dbf
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.prj
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.sbx
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.shp
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.shp.xml
│ Lakes_and_Rivers_Shapefile_NA_Lakes_and_Rivers_data_hydrography_l_rivers_v2.shx
│ Source.txt
│
├───UnitedStates_borders
│ cb_2018_us_state_20m.cpg
│ cb_2018_us_state_20m.shp.ea.iso.xml
│ cb_2018_us_state_20m.shp
│ cb_2018_us_state_20m.shp.iso.xml
│ cb_2018_us_state_20m.prj
│ cb_2018_us_state_20m.dbf
│ cb_2018_us_state_20m.shx
│ README.txt
│
└───Illinois Synthetic Grid
│ Source.txt
│
└───ACTIVSg200
ACTIVSg200.aux
ACTIVSg200.EPC
ACTIVSg200.pwb
ACTIVSg200.pwd
ACTIVSg200.RAW
ACTIVSg200.tsb
ACTIVSg2000.pwd
ACTIVSg200_dynamics.dyd
ACTIVSg200_dynamics.dyr
ACTIVSg200_GIC_data.gic
case_ACTIVSg200.m
contab_ACTIVSg200.m
scenarios_ACTIVSg200.m
This tutorial assumes the use of gitbash or a Unix-like terminal with github command line usage.
- This project utilizes conda to manage environments and ensure consistent results. Download miniconda and ensure you can activate it from your terminal by running
$conda activate- Depending on system configuration, this can be an involved process here is a recommended thread.
- Clone the repository using
$git clone https://github.com/kravitsjacob/water-OPF.git - Download the associated input/output
water-OPF-io-v2.3data here. Place it in the cloned directory. The tree should appear EXACTLY as it does above. - Change to the current working directory using
$cd <insert_path>/water-OPF - Run the analysis by running
$bash run.sh- To keep this project open source, by default this script does not call the simple Matlab converter functions, and the pre-computed outputs are supplied. However, for the sake of transparency, I have included the Matlab code.
The decision tree plotting package dtreeviz and the power system plotting package pandapower.plotting have some dependency issues that prevents dtreeviz from being used if pandapower.plotting is imported. These errors are caughts by main.py to avoid a failing exit code, but the decision trees will not be created. If you want to recreate the decision trees, you must comment line 4 in src/viz.py. Please reach out if you know a workaround!