-### batt_sys_identification
-Battery system identification module. Hosts the class for generating battery system identification parameters
-from experimental data. This module leverages a genetic algorithm to optimize the battery model parameters.
-The battery model is a 2nd order RC Equivalent circuit model (ECM). One can this module to generate custom NMC
-battery parameters by uploading experimental data to the `batt_sys_identification/data` folder and running the module.
-The module will generate a `.csv` file with the battery parameters in the `batt_sys_identification/params` folder.
-The data prototype is shown below. Note that column fields are case-sensitive.
-
-
+### data/elec_rates
+Includes .csv files for electricity Time-of-use (TOU) rates. The input data prototype for electricity rates
+is shown below. User must upload a normal full-year sized data (for 365 days) to avoid
+any errors.
-The module will save a new `.csv` file with an additional field for the corrected open circuit voltage (OCV) values;
-this field (column) will be labelled `ocv_corr` within the new battery data csv, including the existing columns as shown
-in the data prototype above.
+
-Once the battery parameters are generated, they can be used in the `battery_data` folder and `configs/battery.json` can
-be modified so the model runs using the new custom parameters.
+The data required must be in the format shown above. If you upload your own electricity rate data, navigate to `charging_sim/configs/prices.json` and modify the *data_path* field
+to ensure that the *data_path* field matches the uploaded custom price data. The `electricityPrices.py` module will read
+the time-of-use (TOU) price data and sample prices during optimization and simulation. The data should be one full year of TOU rate prices at 15 minute resolution.
+The `electricityPrices.py` module can also help with downscaling the data to 15 minute resolution if the data is at a
+much coarser resolution. The module will save the downscaled data in the `elec_rates` folder.
-### charging_sim
+### data/solar_data
+`solar_data` folder includes solar irradiance data for capturing the effects of environmental conditions on overall system cost. Default
+data for solar irradiance is from the National Solar Radiation Database (NSRDB) for the San Francisco Bay Area.
+The data prototype is from the National Renewable Energy Laboratory (NREL) and is shown below. Note that column fields
+are case-sensitive. If you upload your own solar irradiance data, navigate to `charging_sim/configs/solar.json` and modify
+the *data_path* field to ensure that the *data_path* field matches the uploaded custom solar data.
-This contains the `configs` folder which includes the configuration files for all the relevant modules.Hosts the implementation of the physical modules, including:
+
-`battery.py` - Battery cell module.
+Month labels are indexed from 1 to 12, inclusive; 1 - January, 12 - December. The original data is in hourly resolution.
+The *EV-EcoSim* data prototype is in 15 minute intervals by default, with irradiance oversampled 4 times from hourly
+dataset. The GHI represents the "Global Horizontal Irradiance" in W/m^2, which is the total amount of shortwave radiation
+received from above by a surface horizontal to the ground.
-`batterypack.py` - Battery pack module.
-`batteryAgingSim.py` - Battery aging module.
-`controller.py` - Controller module.
+### batt_sys_identification
+Battery system identification module. Hosts the class for generating battery system identification parameters
+from experimental data. This module leverages a genetic algorithm to optimize the battery model parameters.
+The battery model is a 2nd order RC Equivalent circuit model (ECM). One can use this module to generate custom NMC
+battery parameters by uploading experimental data to the `batt_sys_identification/data` folder and running the module.
+The module will generate a `.csv` file with the battery parameters in the `batt_sys_identification` folder.
+The data prototype is shown below. Note that column fields are case-sensitive.
-`chargingStation.py` - Charging station module.
+
-`electricityPrices.py` - Electricity prices module.
+[//]: # () -`optimization.py` - Optimization module. +[//]: # (
)
-`orchestrator.py` - Simulation orchestrator module.
+[//]: # ( 
)
-`solar.py` - Solar PV module.
+[//]: # ( Equivalent circuit model (ECM) for battery system identification )
-`utils.py` - Hosts utility functions used by some modules.
+[//]: # (
)
-`simulate.py` - Offline DER control optimization for cost minimization (this is run for offline mode (no state feedback)).
+[//]: # ()
-`solar.py` - Solar PV module.
+[//]: # ( ) +The module will save a new `.csv` file with an additional field for the corrected open circuit voltage (OCV) values; +this field (column) will be labelled `ocv_corr` within the new battery data csv, including the existing columns as shown +in the data prototype above. -### DLMODELS +Once the battery parameters are generated, they can be used in the `data/battery_data` folder and `charging_sim/configs/battery.json` can +be modified so the model runs using the new custom parameters. -This includes legacy load forecasts models developed (not needed). +The image below shows the battery model error (MAPE) for the NMC battery model parameters generated from the module for +3 different sample cells over 10 different trials. The model is trained for only 40 generations (or iterations) of the +genetic algorithm with a population of 10. +[//]: # () +
-### elec_rates
+### charging_sim
+This folder encompasses the implementation of the physical modules, including:
+
+ * `battery.py` - Battery cell module.
+ * `batteryAgingSim.py` - Battery aging module.
+ * `batterypack.py` - Battery pack module.
+ * `chargingStation.py` - Charging station module.
+ * `clock.py` - Clock module.
+ * `controller.py` - Controller module.
+ * `electricityPrices.py` - Electricity prices module
+ * `node.py` - Node module (for centralized DER control and optimization).
+ * `optimization.py` - Optimization module.
+ * `orchestrator.py` - Simulation orchestrator module.
+ * `simulate.py` - Offline DER control optimization for cost minimization (this is run for offline mode (no state feedback)).
+ * `solar.py` - Solar PV module.
+ * `transformer.py` - Transformer module.
+ * `utils.py` - Hosts utility functions used by some modules.
+
+There also contains the `configs` folder under the `charging_sim` folder. The `configs` folder which includes the
+configuration files for all the relevant modules, such as battery, transformer, solar, clock modules, etc.
-Includes .csv files for electricity Time-of-use (TOU) rates. The input data prototype for electricity rates
-is shown below. User must upload a normal full-year sized data (for 365 days) to avoid
-any errors.
+### DLMODELS
+This includes legacy load forecasts models developed (not needed).
-
-The data required must be in the format shown above. The `electricityPrices.py` module will read the data and sample
-prices during optimization and simulation. The data should be one full year of TOU rate prices at 15 minute resolution.
-The `electricityPrices.py` module can also help with downscaling the data to 15 minute resolution if the data is at a
-much coarser resolution. The module will save the downscaled data in the `elec_rates` folder.
### feeders
Library of IEEE test feeders and PNNL taxonomy feeders for distribution systems in the GridLAB-D `.glm` format.
IEEE feeders have spot loads specified at primary distribution level. PNNL taxonomy feeders have spot loads specified at
primary or secondary distribution level.
-
### feeder_population
Scripts for populating base feeder models with time-varying loads and resources using the load data in base_load_data.
`feeder_population.py` generates the necessary files for a co-simulation run based on the parameters specified in
-`feeder_population/config.txt`. This module uses the
-Requires residential load data not included in repo (limited access).
-
+`test_cases/{CASE_NAME}/feeder_population/config.txt`. Feeder population requires residential load data not included in repo (limited access) due to proprietary data rights. However, there are
+free versions of Pecan Street data that may be replaced in the `base_load_data` folder; file should be named `data_use.csv`.
-### solar_data
-Includes solar irradiance data for capturing the effects of environmental conditions on overall system cost. Default
-data for solar irradiance is from the National Solar Radiation Database (NSRDB) for the San Francisco Bay Area.
-The data prototype is from the National Renewable Energy Laboratory (NREL) and is shown below. Note that column fields
-are case-sensitive.
-
-
-
-Month labels are indexed from 1 to 12, inclusive; 1 - January, 12 - December. The original data is in hourly resolution.
-The *EV-Ecosim* data prototype is in 15 minute intervals by default, with irradiance oversampled 4 times from hourly
-dataset. The GHI represents the "Global Horizontal Irradiance" in W/m^2, which is the total amount of shortwave radiation
-received from above by a surface horizontal to the ground.
-
### test_cases
#### Co-simulation cases.
`base_case`- Reads voltage from GridLAB-D and writes power injections at each timestep (no EV charging or DER).
-
-`rlsf` - base_case plus implements a recursive least squares filter to estimate network model online (not used)
-
`battery` - base_case plus transformer thermal model plus DER integration (included battery and solar).
-`transformer` - base_case plus simulation of transformer thermal model for each transformer in GridLAB-D model (not used).
-
-
-
### analysis
-
Scripts for plotting and analysis of co-simulation results. Includes post optimization and simulation cost
calculation modules and voltage impacts on the distribution grid.
@@ -144,32 +170,60 @@ voltage violations per ANSI C84.1. The file also generates voltage distribution
SIMULATION_FOLDER variable which is the string of the path where the powerflow simulation output voltages at each node
exist.
-`load_post_opt_costs.py` - This module is calculates the levelized cost of energy and populates into tables/cost matrices, which are saved in the
+`load_post_opt_costs.py` - This module calculates the levelized cost of energy and populates into tables/cost matrices, which are saved in the
respective files and folders. The module also generates plots for the cost analysis.
`cost_analysis.py` - This module contains the `CostEstimator` class, which estimates the cost of the different grid and DER components
from the simulation.
-
## How to run
-Create a new environment using `conda env create --name