This package gathers and builds demand, hydro, solar, and wind profiles. The profiles are needed to run scenarios on the U.S. electrical grid. PreREISE is part of a set of packages representing Breakthrough Energy's power system model.

More information regarding the installation of the model as well as the contribution guide can be found here.

Here are the instructions to install the PreREISE package. We strongly recommend that you pick one of the following options.

If not already done, install pipenv by following the instructions on their webpage. Then run:

pipenv sync
pipenv shell

in the root folder of the package. The first command will create a virtual environment and install the dependencies. The second command will activate the environment.

First create an environment using venv (more details here). Note that venv is included in the Python standard library and requires no additional installation. Then, activate your environment and run:

pip install -r requirements.txt

in the root folder of the package.

Whatever method you choose, if you wish to access the modules located in PreREISE from anywhere on your machine, do:

pip install .

in the root folder of your package or alternatively, setup the PYTHONPATH global variable.

This module aims at gathering the required data for building the profiles

We call the time series profiles directly generated from mathematical/heuristic models without any scaling (as we do in different scenarios) as 'raw' profiles.

The name of a raw profile in the simulation framework must be of the form:

[type]_v[D],

where type specifies the profile type ('demand', 'hydro', 'solar', 'wind') and D refers to the date the file has been created. Month (3 letters, e.g., Jan) and year (4 digits, e.g., 2021) are used for the date. If two or more profiles of the same type are created the same month of the same year, the different versions will be differentiated using a lowercase letter added right after the year, e.g, demand_vJan2021.csv (first), demand_vJan2021b.csv (second), demand_vJan2021c.csv (third) and so forth.

RAP (Rapid Refresh) is the continental-scale NOAA hourly-updated assimilation/modeling system operational at the National Centers for Environmental Prediction (NCEP). RAP covers North America and is comprised primarily of a numerical weather model and an analysis system to initialize that model. RAP provides, every hour ranging from May 2012 to date, the U and V components of the wind speed at 80 meter above ground on a 13x13 square kilometer resolution grid every hour. Data can be retrieved using the NetCDF Subset Service. Information on this interface is described here. Note that the dataset is incomplete and, consequently, missing entries need to be imputed.

Once the U and V components of the wind are converted to a non-directional wind speed magnitude, this speed is converted to power using wind turbine power curves. Since real wind farms are not currently mapped to TAMU network farms, a capacity-weighted average wind turbine power curve is created for each state based on the turbine types reported in EIA Form 860. The wind turbine curve for each real wind farm is looked up from a database of curves (or the IEC class 2 power curve provided by NREL in the WIND Toolkit documentation) is used for turbines without curves in the database), and scaled from the real hub heights to 80m hub heights using an alpha of 0.15. These height-scaled, turbine-specific curves are averaged to obtain a state curve translating wind speed to normalized power. States without wind farms in EIA Form 860 are represented by the IEC class 2 power curve.

Each turbine curve represents the instantaneous power from a single turbine for a given wind speed. To account for spatio-temporal variations in wind speed (i.e. an hourly average wind speed that varies through the hour, and a point-specific wind speed that varies throughout the wind farm), a distribution is used: a normal distribution with standard deviation of 40% of the average wind speed. This distribution tends to boost the power produced at lower wind speeds (since the power curve in this region is convex) and lower the power produced at higher wind speeds (since the power curve in this region is concave as the turbine tops out, and shuts down at higher wind speeds). This tracks with the wind-farm level data shown in NREL's validation report.

Check out the rap_demo.ipynb notebook for demo.

The Gridded Atmospheric WIND (Wind Integration National Dataset) Toolkit provides 1-hour resolution irradiance data for 7 years, ranging from 2007 to 2013, on a uniform 2x2 square kilometer grid that covers the continental U.S., the Baja Peninsula, and parts of the Pacific and Atlantic oceans. Data can be accessed using the Highly Scalable Data Service. NREL wrote example notebooks that demonstrate how to access the data.

Power output is estimated using a simple normalization procedure. For each solar plant location the hourly Global Horizontal Irradiance (GHI) is divided by the maximum GHI over the period considered and multiplied by the capacity of the plant. This procedure is referred to as naive since it only accounts for the plant capacity. Note that other factors can possibly affect the conversion from solar radiation at ground to power such as the temperature at the site as well as many system configuration including tracking technology.

Check out the ga_wind_demo.ipynb notebook for demo.

NSRDB (National Solar Radiation Database) provides 1-hour resolution solar radiation data, ranging from 1998 to 2016, for the entire U.S. and a growing list of international locations on a 4x4 square kilometer grid. Data can be accessed via an API. Note that the Physical Solar Model v3 is used.

An API key is required to access and use the above databases. Get your own API key here.

Here, the power output can be estimated using the previously presented naive method or a more sophisticated one. The latter uses the System Adviser Model (SAM) developed by NREL. The developer tools for creating renewable energy system models can be downloaded here. Irradiance data along with other meteorological parameters must first be retrieved from NSRDB for each site. This information are then fed to the SAM Simulation Core (SCC) and the power output is retrieved. The SSC reflect the technology used: photovoltaic (PV), solar water heating and concentrating solar power (CSP). The PVWatts v7 model is used for all the solar plants in the grid. The system size (in DC units) and the array type (fixed open rack, backtracked, 1-axis and 2-axis) is set for each solar plant whereas a unique value of 1.25 is used for the DC to AC ratio (see article from EIA on inverter loading ratios here). Otherwise, all other input parameters of the PVWatts model are set to their default values. EIA reports in form 860 the array type used by each solar PV plant. For each plant in our network, the array type is a combination of the three technology that is calculated using the capacity weighted average of the array type of all the plants in EIA Form 860 that are located in the same state. If no plants are reported in EIA Form 860 for a particular state we then use the information of all the plants belonging to the same interconnect.

The naive and the SAM methods are used in the nsrdb_naive_demo.ipynb and nsrdb_sam_demo.ipynb demo notebooks, respectively.

EIA (Energy Information Administration) published monthly capacity factors for hydro plants across the country. This dataset (available here) is used to produce a profile for each hydro plant in the grid. Note that we are using the same set of capacity factor independently of the geographical location of the plant. As a result, the profile of all the hydro plants in the grid will have the same shape. Only the power output will differ (the scale of the profile).

Check out the hydro_v1_demo.ipynb notebook for demo.

Hydro hourly profile v2 for Western consists of two components: profile shape to capture the hourly variance and monthly total generation to capture the historical summation. EIA published monthly total generation for each hydro plant by state in form EIA 923 (available here).

Given that Washington state has the most hydro generation in the Western Interconnection, we use the aggregated generation of the top 20 US Army Corps managed hydro dams in Northwestern US (primarily in WA) as the shape of this hydro profile v2 for all the states in Western except for California and Wyoming. The data is obtained via US Army Corps of Engineers Northwestern Division DataQuery Tool 2.0.

Observed from system daily outlook of California ISO (available here), the hydro generation profile follows closely to the net demand profile. Thus, we construct the shape curve of hydro profile v2 in CA using the net demand profile of CA in the base case scenario (Scenario 87). As for Wyoming, due to the small name plate capacities of the hydro generators, we simply apply a constant shape curve to avoid peak-hour violation of maximum capacity.

The final hydro profile v2 is built by scaling the corresponding shape curves based on the monthly total generation record in each state, then decomposing into plant-level profiles proportional to the generator capacities. Check out the western_hydro_v2_demo.ipynb notebook for demo.

The Electric Reliability Council of Texas (ERCOT) published actual generation by fuel type for each 15 minute settlement interval. For details, see the fuel mix report available here. Given this time-series historical data, similarly with the western case, the final hydro profile v2 for Texas is constructed by decomposing the total hydro generation profile proportional to the generator capacities. Check out the texas_hydro_v2_demo.ipynb notebook for demo.

This methodology is designed to generate hourly plant level hydro profile for Eastern, i.e. eastern hydro v3, which has following features:

• Pumped storage hydro (HPS) and conventional hydro (HYC) are handled separately.
• Historical hourly hydro profiles of four ISOs (ISONE, NYISO, PJM and SPP), are used directly as regional total hydro profiles to start with.
• For the rest of areas in Eastern Interconnect, which are not covered by the four ISOs, the hydro profile is generated in the same way as western hydro profile v2.

Note that usa hydro v3 is the concatenation of eastern hydro v3, western hydro v2 and texas hydro v2

HPS profile for each plant in hps_plants_eastern.xlsx, ‘all_plantIDs’ sheet, is generated based on the deterministic model presented in ‘profile’ sheet. The HPS profile model is designed based on local time. The plant level HPS profiles are generated considering the local time and daylight-saving schedule of the locations of the corresponding buses. However, during the sanity check afterwards, it turns out that this model overestimates the HPS output, the final profile is further scaled by 0.65 to be in agreement with values reported in EIA 923.

The buses of the HYC plants were associated to five regions (ISONE, NYISO, PJM, SPP and the area uncovered by the aforementioned regions) using the same methodology employed for eastern demand v5. Check Demand Data section below for details. For HYC plants locating inside the territories of the four ISOs, we decompose the total historical profile of each ISO into plant -level profiles proportional to the generator capacities. The time series profile data of each ISO is obtained via their official websites: ISONE, NYISO, PJM, SPP.

For the rest of HYC plants lying outside the four ISO regions, we apply the same methodology that is used to generate hydro profile of California in Western hydro v2 (see details in the above subsection). The net demand profile is obtained from Eastern base case scenario (Scenario 397). The monthly net generations for a state is split proportional to the total capacities of the generators if the corresponding state is covered partially by the four ISOs mentioned above.

The final Eastern hydro profile v3 is built by stitching the three parts together: pumped storage hydro profiles, conventional hydro profiles within the four ISO regions and conventional hydro profiles in the rest areas . Check out eastern_hydro_v3_demo.ipynb notebook for demo.

Eastern V6

The two BAs, SPP and MISO, consist of big areas from north to south. It is unlikely that the whole area covered by these big BAs have the same hourly profile shape. Eastern demand v6 addresses the issue by further splitting these two BAs into subareas. We replace overall MISO and SPP demand with subarea demand obtained directly from contact at the BAs. We use overall numbers from EIA and use the fractional subarea demand as the basis to split the MISO and SPP demand.

The hourly subarea demand profiles for SPP and MISO is reported on their official websites respectively (check SPP hourly load, MISO hourly load). The geographical definitions of these subareas are obtained directly from contact at the BAs via either customer service or internal request management system (check RMS). Given the geographical definitions, i.e. list of counties for each subarea, each bus is mapped to the subarea it belongs to.

The overall procedure is similar to v5 except we generate subarea mapping as well as prepare subarea demand from files.

Eastern V5

Demand data are obtained from EIA, to whom Balancing Authorities have submitted their data. The data can be obtained using an API as demonstrated in eastern_demand_v5_demo.ipynb notebook.

Module get_eia_data contains functions that converts the data into data frames for further processing, as performed in eastern_demand_v5_demo.ipynb.

To output the demand profile, cleaning steps were applied to the EIA data for Eastern V5. We used adjacent demand data to fill missing values using a series of rules:

1. Monday: look forward one day
2. Tuesday - Thursday: average of look forward one day and look back one day
3. Friday: look back one day
4. Saturday: look forward one day
5. Sunday: look back one day

If data is still missing after applying the above rules, week ahead and week behind data is used:

1. Monday: look forward two days
2. Tuesday: look forward two days
3. Wednesday: average of look forward two days and look back two days
4. Thursday: look back two days
5. Friday: look back two days
6. Saturday - Sun: average of look back one week and look forward one week

If data is still missing after applying the above rules, week ahead and week behind data is used:

1. Monday - Sunday: average of look back one week and look forward one week

The next step was outlier detection. The underlying physical rationale is that demand changes are mostly driven by weather temperature changes (first or higher order), and thermal mass limits the rate at which demand values can change. Outliers were detected using a z-score threshold value of 3. These outliers are then replaced by linear interpolation.

The BA counts were then distributed across each region where the BA operates, using the region populations as weights. For example, if a BA operates in both WA and OR, the counts for WA are weighted by the fraction of the total counts in WA relative to the total population of WA and OR.

Lastly, buses were mapped to counties. This step requires an input data file that stores the list of counties in each BA area territory. Some data cleaning was necessary to deal with inconsistent county names. We also implemented a check if there are buses where BA is empty/not found, which is due to bus being outside of United States. These were fixed manually by assigning the bus to the nearest county.

Legacy Description

Demand data are obtained from EIA, to whom Balancing Authorities have submitted their data. The data can be obtained either by direct download from their database using an API or by download of Excel spreadsheets. An API key is required for the API download and this key can be obtained by a user by registering at https://www.eia.gov/opendata/.

The direct download currently contains only published demand data. The Excel spreadsheets include original and imputed demand data, as well as results of various data quality checks done by EIA. Documentation about the dataset can be found here. Excel spreadsheets can be downloaded by clicking on the links in page 9 (Table of all US and foreign connected balancing authorities).

Module get_eia_data contains functions that converts the data into data frames for further processing.

To test EIA download (This requires an EIA API key):

from prereise.gather.demanddata.eia.tests import test_get_eia_data

test_get_eia_data.test_eia_download()

To test EIA download from Excel:

from prereise.gather.demanddata.eia.tests import test_get_eia_data

test_get_eia_data.test_from_excel()

The assemble_ba_from_excel_demo.ipynb notebook illustrates usage.

To output the demand profile, cleaning steps were applied to the EIA data: 1) missing data imputation - the EIA method was used, i.e., EIA published data was used; beyond this, NA's were converted to float zeros; 2) missing hours were added.

The BA counts were then distributed across each region where the BA operates, using the region populations as weights. For example, if a BA operates in both WA and OR, the counts for WA are weighted by the fraction of the total counts in WA relative to the total population of WA and OR.

The next step consist in detecting outliers by looking for large changes in the slope of the demand data. The underlying physical rationale is that demand changes are mostly driven by weather temperature changes (first or higher order), and thermal mass limits the rate at which demand values can change. By looking at the slope of demand data, it is seen that the slope distribution is normally distributed, and outliers can be easily found by imposing a z-score threshold value of 3. These outliers are then replaced by linear interpolation.

To test outlier detection, use:

from prereise.gather.demanddata.eia.tests import test_clean_data

test_clean_data.test_slope_interpolate()

The ba_anomaly_detection_demo.ipynb notebook illustrates usage.

The National Renewable Energy Laboratory (NREL) has developed the Electrification Futures Study (EFS) to project and study future sectoral demand changes as a result of impending widespread electrification. As a part of the EFS, NREL has published multiple reports (dating back to 2017) that describe their process for projecting demand-side growth and provide analysis of their preliminary results; all of NREL's published EFS reports can be found here. Accompanying their reports, NREL has published data sets that include hourly load profiles that were developed using processes described in the EFS reports. These hourly load profiles represent state-by-state end-use electricity demand across four sectors (Transportation, Residential Buildings, Commercial Buildings, and Industry) for three electrification scenarios (Reference, Medium, and High) and three levels of technology advancements (Slow, Moderate, and Rapid). Load profiles are provided for six separate years: 2018, 2020, 2024, 2030, 2040, and 2050. The base demand data sets and further accompanying information can be found here. In addition to demand growth projections, NREL has also published data sets that include hourly profiles for flexible load. These data sets indicate the amount of demand that is considered to be flexible, as determined through the EFS. The flexibility demand profiles consider two scenarios of flexibility (Base and Enhanced), in addition to the classifications for each sector, electrification scenario, technology advancement, and year. The flexible demand data sets and further accompanying information can be found here.

Widespread electrification can have a large impact on future power system planning decisions and operation. While increased electricity demand can have obvious implications for generation and transmission capacity planning, new electrified demand (e.g., electric vehicles, air source heat pumps, and heat pump water heaters) offers large amounts of potential operational flexibility to grid operators. This flexibility by demand-side resources can result in demand shifting from times of peak demand to times of peak renewable generation, presenting the opportunity to defer generation and transmission capacity upgrades. To help electrification impacts be considered properly, this package has the ability to access NREL's EFS demand data sets. Currently, users can access the base demand profiles, allowing the impacts of demand growth due to vast electrification to be explored. Users can also access the EFS flexible demand profiles, though integration of demand-side flexibility modeling to the rest of Breakthrough Energy Sciences' production cost model will be included in a future release.

The EFS demand data sets are stored on the 'NREL Data Catalog' in .zip files. The 'NREL Data Catalog' contains nine .zip files, one for each combination of electrification scenario and technology advancement. The .csv file compressed within the .zip file contains the sectoral demand for each state and year. This data can be downloaded and extracted using the following:

from prereise.gather.demanddata.nrel_efs.get_efs_data import download_demand_data

download_demand_data(es, ta, fpath, sz_path)

where es is the set of electrification scenarios to be downloaded, ta is the set of technology advancements to be downloaded, fpath is the file path to which the NREL EFS data will be downloaded, and sz_path is the file path to which 7-Zip is located for Windows users. es and ta default to downloading each of the electrification scenarios and technology advancements, respectively. fpath defaults to downloading the EFS data to the current directory. sz_path defaults to the default installation location for 7-Zip.

The EFS flexibility data sets are also stored on the 'NREL Data Catalog' in .zip files. The flexibility data is stored differently from the base demand data, with flexibility data stored in three separate .zip files, one for each electrification scenario. The .csv file compressed within the .zip file contains the sectoral flexibility for each state, year, technology advancement, and flexibility scenario. This data can be downloaded and extracted using the following:

from prereise.gather.demanddata.nrel_efs.get_efs_data import download_flexibility_data

download_flexibility_data(es, fpath, sz_path)

where es is the set of electrification scenarios to be downloaded, fpath is the file path to which the NREL EFS data will be downloaded, and sz_path is the file path to which 7-Zip is located for Windows users. es defaults to downloading each of the electrification scenarios. fpath defaults to downloading the EFS data to the current directory. sz_path defaults to the default installation location for 7-Zip.

Although downloading the .zip files is a simple task, extracting the .csv files using Python is more challenging. The .zip files stored on the 'NREL Data Catalog' are compressed using a compression method (Deflate64, or compression type 9) that is not currently supported by zipfile, a popular Python package for working with .zip files. To extract the .csv files in an automated fashion, it is necessary to use either the Command Prompt or the Terminal, depending on the user's operating system. For machines running macOS or Linux, the Terminal's extraction tools are capable of accessing the .csv file. However, Windows machines are unable to extract the .csv files using the Command Prompt's extraction tools due to the compression method. For Windows users with 7-Zip, the .csv file can be extracted, so long as 7-Zip is installed in the default location. If none of these methods succeed in extracting the .csv file, then users can extract the file manually by going to the .zip file's location and using their extraction tool of choice. For instance, even though Windows' Command Prompt extraction tools do not work, the native extraction tool built into Windows' File Explorer does work.

The EFS demand data for a given electrification scenario and technology advancement is provided for each sector and each year. Although the sectoral demand is eventually aggregated for a given year (see next subsection), splitting the demand by sector can be useful, especially for users that may only want to use a subset of the EFS sectoral demand (perhaps to pair with their own sectoral demand data). EFS demand data for a single electrification scenario and technology advancement pair can be split by sector as follows:

from prereise.gather.demanddata.nrel_efs.get_efs_data import partition_demand_by_sector

sect_dem = partition_demand_by_sector(es, ta, year, sect, fpath, save)

where es is a string describing a single electrification scenario, ta is a string describing a single technology advancement, year is an integer describing the included year, sect is a set of the sectors to include, fpath is the file path where the demand data might be located and to where the sectoral demand will be saved (if desired), and save is a boolean that indicates whether the sectoral demand should be saved. sect defaults to including all the sectors (i.e., all sectoral demand is retained). fpath defaults to the current directory. save defaults to False (i.e., each sectoral demand DataFrame is not saved). partition_demand_by_sector returns sect_dem, which is a dictionary of DataFrames, where each DataFrame contains the demand data for one sector.

The EFS flexibility data for a given electrification scenario is provided for each sector, year, technology advancement, and flexibility scenario. It is useful to split the flexibility data by sector for a specific year, technology advancement, and flexibility scenario. It is useful to split flexibility data by sector because different sectors may have different operational constraints (e.g., duration between demand curtailment and recovery, directionality of load shift). EFS flexibility data can be split by sector as follows:

from prereise.gather.demanddata.nrel_efs.get_efs_data import partition_flexibility_by_sector

sect_dem = partition_flexibility_by_sector(es, ta, flex, year, sect, fpath, save)

where es is a string describing a single electrification scenario, ta is a string describing a single technology advancement, flex is a string describing a single flexibility scenario, year is an integer describing the included year, sect is a set of the sectors to include, fpath is the file path where the flexibility data might be located and to where the flexibility demand will be saved (if desired), and save is a boolean that indicates whether the sectoral flexibility data should be saved. sect defaults to including all the sectors (i.e., all sectoral flexibility is retained). fpath defaults to the current directory. save defaults to False (i.e., each sectoral flexibility DataFrame is not saved). partition_flexibility_by_sector returns sect_flex, which is a dictionary of DataFrames, where each DataFrame contains the flexibility data for one sector.

partition_demand_by_sector and partition_flexibility_by_sector check the file path provided by fpath to determine if the data is already downloaded and extracted. If the data is not present as a .csv file, then partition_demand_by_sector and partition_flexibility_by_sector use download_demand_data and download_flexibility_data, respectively, to obtain the specified .csv file. If the automated approach is not able to extract the .csv file, partition_demand_by_sector and partition_flexibility_by_sector will exit with an error, and the user will need to manually use their extraction method of choice (as was discussed in the prior subsection). If manual extraction is required, the user can simply run partition_demand_by_sector and partition_flexibility_by_sector again, making sure that fpath points to the appropriate location where the .csv file is saved.

For use in the grid model developed by Breakthrough Energy Sciences, all sectoral demand must be aggregated within each location for each hour (i.e., only one demand data point per location per hour). Thus, the sectoral demand must be aggregated together to obtain a single demand value for each location and each hour. Aggregating sectoral demand can be accomplished as follows:

from prereise.gather.demanddata.nrel_efs.aggregate_demand import combine_efs_demand

agg_dem = combine_efs_demand(efs_dem, non_efs_dem, save)

where efs_dem is a dictionary of different sectoral demand DataFrames pertaining to the EFS (this input is intended to be the output of partition_demand_by_sector), non_efs_dem is a list of different sectoral demand DataFrames that are independent of the EFS (this input is intended to be the output of access_non_efs_demand, which is addressed below), and save is a string representing the desired file path and file name to which the aggregated demand will be saved as a .csv file. Both efs_dem and non_efs_dem default to None, allowing the user to determine how much sectoral demand of each type (EFS and non-EFS) they would like to include; note that failing to specify any sectoral demand will raise an error. save defaults to None, indicating that the aggregate demand should not be saved. combine_efs_demand returns agg_dem, which is a DataFrame of the aggregate demand data.

The access_non_efs_demand function allows sectoral demand that is not associated with the EFS to be used. access_non_efs_demand loads locally-stored sectoral demand data, checks that it is formatted appropriately, and returns a list of sectoral demand DataFrames that can be fed into combine_efs_demand. Preparing non-EFS sectoral demand is accomplished as follows:

from prereise.gather.demanddata.nrel_efs.aggregate_demand import access_non_efs_demand

sect_dem = access_non_efs_demand(dem_paths)

where dem_paths is a list of file paths that point to the .csv files of locally-stored sectoral demand data. access_non_efs_demand is intended to be used on sectoral demand data that is not related to the EFS; all sectoral demand data that is related to the EFS should be handled using partition_demand_by_sector. Note that it will be incumbent upon the user to account for all sectors when building the aggregate demand profiles (i.e., users will need to ensure that specific sectors are not excluded or double-counted).

For use in the grid model developed by Breakthrough Energy Sciences, the state-level demand provided by the EFS must be mapped to the appropriate load zones. Breakthrough Energy Sciences defines distinct load zones for each state. For smaller states that are completely contained within a single interconnection, the load zone might very well be equal to the whole state. However, large states (e.g., California and New York), states that are in multiple interconnections (e.g., Montana and New Mexico), and states with a combination of these two characteristics (e.g., Texas) may have multiple load zones. State-level demand is mapped to the various load zones according to the percentage of a state's population that resides in a particular load zone. This mapping can be accomplished as follows:

from prereise.gather.demanddata.nrel_efs.map_states import decompose_demand_profile_by_state_to_loadzone

df_lz = decompose_demand_profile_by_state_to_loadzone(df, save)

where df is a DataFrame of the hourly demand data for each state in the contiguous U.S. (intended to be the output of combine_efs_demand or the components that are output from partition_flexibility_by_sector) and save is a string representing the desired file path and file name to which the demand will be saved as a .csv file. save defaults to None, indicating that the demand should not be saved. decompose_demand_profile_by_state_to_loadzone returns df_lz, which is a DataFrame of the demand data mapped to each load zone.

To demonstrate the work flow of these modules, this subsection presents an example of obtaining the EFS demand data and subsequently preparing it for use in Breakthrough Energy Sciences' grid model. In this example, EFS demand data under the 'Reference' electrification scenario and the 'Slow' technology advancement are acquired for the year 2030. After mapping the state-level demand to the different load zones, the aggregate demand is saved as a .csv file to the current working directory. This example is implemented as follows:

from prereise.gather.demanddata.nrel_efs.aggregate_demand import combine_efs_demand
from prereise.gather.demanddata.nrel_efs.get_efs_data import partition_demand_by_sector
from prereise.gather.demanddata.nrel_efs.map_states import decompose_demand_profile_by_state_to_loadzone

sect_dem = partition_demand_by_sector(es="Reference", ta="Slow", year=2030, fpath="")
agg_dem = combine_efs_demand(efs_dem=sect_dem)
agg_dem_lz = decompose_demand_profile_by_state_to_loadzone(df=agg_dem, save="EFS_Demand_Reference_Slow_2030.csv")

This subsection presents an example of obtaining the EFS flexibility data and subsequently preparing it for use in Breakthrough Energy Sciences' grid model. In this example, EFS flexibility data under the 'Reference' electrification scenario, 'Slow' technology advancement, and 'Base' flexibility scenario are acquired for the year 2030. The state-level sectoral flexibility is mapped to the different load zones, however the different flexibility data are not saved as .csv files as was done in the prior example. This example is implemented as follows:

from prereise.gather.demanddata.nrel_efs.get_efs_data import partition_flexibility_by_sector
from prereise.gather.demanddata.nrel_efs.map_states import decompose_demand_profile_by_state_to_loadzone

sect_flex = partition_flexibility_by_sector(es="Reference", ta="Slow", flex="Base", year=2030, fpath="")
sect_flex_lz = {k: decompose_demand_profile_by_state_to_loadzone(df=v) for k, v in sect_flex.items()}