# EIA Form 930 – Hourly and Daily Balancing Authority Operations Report
| Source URL | [https://www.eia.gov/electricity/gridmonitor/](https://www.eia.gov/electricity/gridmonitor/) |
|---------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Source Description | The EIA Form 930 provides hourly demand and generation statistics by balancing area (or sub-balancing area in the case of some larger areas). These statistics include a breakdown by energy source (coal, gas, hydro, wind, solar, etc.) as well as interchange between the balancing areas, including international exchanges with Canada and Mexico. |
| Respondents | All entities in the contiguous United States that are listed in the North American Electric Reliability Corporation’s (NERC) Compliance Registry as a balancing authority. |
| Source Format | Comma separated values (CSV) |
| Download Size | 409 MB |
| Temporal Coverage | 2015half2-2026half2 |
| PUDL Code | `eia930` |
| Unprocessed Source Data Archive | [10.5281/zenodo.10840077](https://doi.org/10.5281/zenodo.10840077) |
| Issues | [Open EIA Form 930 – Hourly and Daily Balancing Authority Operations Report issues](https://github.com/catalyst-cooperative/pudl/issues?utf8=%E2%9C%93&q=is%3Aissue+is%3Aopen+label%3Aeia930) |
## PUDL Database Tables
Clicking on the links will show you a description of the table as well as the names and
descriptions of each of its fields. Due to the size of the EIA-930 hourly tables we only
publish them as Parquet files, which are not browsable online, but they can be
downloaded directly from [Cloud Storage](../data_access.md#access-cloud) via the links embedded in the
[PUDL Data Dictionary](../data_dictionaries/pudl_db.md) or accessed on [Kaggle](../data_access.md#access-kaggle).
| Data Dictionary | Browse Online |
|--------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| [core_eia930_\_hourly_interchange](../data_dictionaries/pudl_db.md#core-eia930-hourly-interchange) | [https://data.catalyst.coop/preview/pudl/core_eia930_\_hourly_interchange](https://data.catalyst.coop/preview/pudl/core_eia930__hourly_interchange) |
| [core_eia930_\_hourly_net_generation_by_energy_source](../data_dictionaries/pudl_db.md#core-eia930-hourly-net-generation-by-energy-source) | [https://data.catalyst.coop/preview/pudl/core_eia930_\_hourly_net_generation_by_energy_source](https://data.catalyst.coop/preview/pudl/core_eia930__hourly_net_generation_by_energy_source) |
| [core_eia930_\_hourly_operations](../data_dictionaries/pudl_db.md#core-eia930-hourly-operations) | [https://data.catalyst.coop/preview/pudl/core_eia930_\_hourly_operations](https://data.catalyst.coop/preview/pudl/core_eia930__hourly_operations) |
| [core_eia930_\_hourly_subregion_demand](../data_dictionaries/pudl_db.md#core-eia930-hourly-subregion-demand) | [https://data.catalyst.coop/preview/pudl/core_eia930_\_hourly_subregion_demand](https://data.catalyst.coop/preview/pudl/core_eia930__hourly_subregion_demand) |
| [out_eia930_\_hourly_aggregated_demand](../data_dictionaries/pudl_db.md#out-eia930-hourly-aggregated-demand) | [https://data.catalyst.coop/preview/pudl/out_eia930_\_hourly_aggregated_demand](https://data.catalyst.coop/preview/pudl/out_eia930__hourly_aggregated_demand) |
| [out_eia930_\_hourly_operations](../data_dictionaries/pudl_db.md#out-eia930-hourly-operations) | [https://data.catalyst.coop/preview/pudl/out_eia930_\_hourly_operations](https://data.catalyst.coop/preview/pudl/out_eia930__hourly_operations) |
| [out_eia930_\_hourly_subregion_demand](../data_dictionaries/pudl_db.md#out-eia930-hourly-subregion-demand) | [https://data.catalyst.coop/preview/pudl/out_eia930_\_hourly_subregion_demand](https://data.catalyst.coop/preview/pudl/out_eia930__hourly_subregion_demand) |
## Background
The EIA Form 930 provides hourly operating data about the high-voltage bulk electric
power grid in the Lower 48 states. See the [official EIA-930 documentation](https://www.eia.gov/electricity/gridmonitor/about) for additional details.
The EIA-930 6-month consolidated CSVs contain the following hourly data elements:
* Demand (MWh)
* Demand by subregion (available for select BAs beginning 2018-07-01)
* Day ahead demand forecasts
* Net generation (MWh)
* Net generation by energy source (beginning 2018-07-01)
* Total net interchange (MWh)
* Interchange with adjacent BAs (MWh)
Additional CO2 emissions data is available through the per-BA Excel spreadsheet data,
**but these spreadsheet data have not (yet) been integrated into PUDL**:
* Estimated total CO2 emissions (beginning 2018-07-01)
* Estimated CO2 emissions by fuel type (beginning 2018-07-01)
* Estimated CO2 emissions for electricity imports or exports (beginning 2018-07-01)
### Demand and demand by subregion
Demand is a calculated value representing the amount of electricity load within a BA’s
electric system. A BA derives its demand value by taking the total metered net
electricity generation within its electric system and subtracting the total metered net
electricity interchange occurring between the BA and its neighboring BAs.
For select BAs, demand by subregion data provide demand at a further level of geographic
disaggregation (for example, load zones, weather zones, operating company areas, local
BAs, etc.) within a BA’s electric system. A BA’s reporting demand by subregion section
below provides more information on subregions. A mapping of BA codes to their subregions
can be found in [core_eia_\_codes_balancing_authority_subregions](../data_dictionaries/pudl_db.md#core-eia-codes-balancing-authority-subregions)
### Day-ahead demand forecast
Each BA produces a day-ahead electricity demand forecast for every hour of the next day.
These forecasts help BAs plan for and coordinate the reliable operation of their
electric system.
### Net generation
Net generation represents the metered output of electric generating units in a BA’s
electric system. This generation only includes generating units that are managed by a BA
or whose operations are visible to a BA.
Generators on the distribution system—both large-scale resources and small-scale
distributed resources, such as rooftop solar photovoltaic (PV) systems—are typically not
included.
BAs report generation from dual-fuel (switching from one fuel to another) and
multiple-fuel (using multiple fuels simultaneously) generators under the actual energy
source used, if known, and under the generator’s primary energy source, if not known.
To maintain generator confidentiality, generation may sometimes be reported in the Other
category if too few generators are reported for a particular energy source category.
In 2024Q3 EIA started using different and more granular energy source categories,
particularly for renewable energy sources. The new categories differentiate between wind
and solar plants that do and don’t have energy storage, and split out pumped hydro from
hydro without storage. They also added new battery storage and geothermal categories.
In some of our electricity publications, we report generation from all utility-scale
generating units in the United States. BAs only meter generating units that are from a
subset of all utility-scale generating units. As a result, when hourly generation from
the EIA-930 data is aggregated to monthly or annual levels, the results will be lower
than monthly and annual aggregations in other electricity publications.
### Total interchange and interchange with adjacent BAs
Interchange is the net metered tie line flow from one BA to another directly
interconnected BA. Total net interchange is the net sum of all interchange occurring
between a BA and its directly interconnected neighboring BAs.
Negative interchange values indicate net inflows, and positive interchange values
indicate net outflows.
### Download additional documentation
* [`Eia930 Instructions 2026-07-31 (PDF)`](eia930/eia930_instructions_2026-07-31.pdf)
### Data available through PUDL
With the exception of the hourly CO2 emisisons data, all of the data mentioned above is
available through PUDL (demand, day ahead demand forecasts, demand by subregion, net
generation, net generation by energy source, total net interchange, and interchange by
adjacent BA)
### Who submits this data?
The EIA-930 data is collected from electricity balancing authorities (BAs) that operate
the grid. Balancing authorities are a functional role defined by the [North American
Electric Reliability Corporation](https://www.nerc.com) (NERC), and are mainly
responsible for balancing electricity supply, demand, and interchange on their electric
systems in real time.
### What does the original data look like?
EIA publishes several versions of the Form 930:
* Bulk CSVs each of which cover six months of data for all BAs (**this is the data
integrated into PUDL**)
* Per-BA Excel spreadsheets with additional CO2 emissions data and other detailed
information about outlying values (not yet used by PUDL).
* A REST API.
## Notable Irregularities
Be sure to check out [the official EIA-930 documentation](https://www.eia.gov/electricity/gridmonitor/about).
EIA publishes the original raw data as it is supplied by the balancing authorities, and
it includes many irregularities including oulying and missing values, which they attempt
to manage in their aggregated (daily or nationwide) data, but which are present in the
original reported data.
For the EIA-930 output tables published in PUDL, in addition to the original reported
electricity demand, we provide a column with anomalous and missing values replaced by
imputed values so the data can be more easily used in capacity expansion modeling
and other applications where having complete and plausible (if not exactly correct)
values is desirable. For an overview of that process and links to additional references,
see [Timeseries Imputation](../methodology/timeseries_imputation.md)
For a visual exploration of the imputed values and comparison with other imputations,
check out the following notebooks on Kaggle:
* [EIA-930 Sanity Checks](https://www.kaggle.com/code/catalystcooperative/03-eia-930-sanity-checks)
* [PUDL Imputed Electricity Demand](https://www.kaggle.com/code/catalystcooperative/06-pudl-imputed-electricity-demand)
### Changes in energy source granularity over time
In the Q2 2025 data release, imputed and adjusted columns for the “unknown” energy
source were added to the raw dataset, so that tracking “unknown” behaves like
tracking any other energy source. This change was backdated as zero to all years.
In the Q1 2025 data release, the wind, solar, and hydro power energy source columns
were removed from the raw dataset, and new columns for multiple different kinds of
wind, solar, hydro power, and energy storage were added. The change takes effect
starting in 2024half2. Analyses that cross this temporal boundary will need to
aggregate in order to compare e.g. 2025 solar with 2023 solar.
| Removed in Q1 2025 release | Added in Q1 2025 release |
|------------------------------|----------------------------------------------------------------------------------------------------|
| | `battery_storage` |
| | `geothermal` |
| `hydro` | `hydro_excluding_pumped_storage`
`pumped_storage`
|
| | `other_energy_storage` |
| `solar` | `solar_w_integrated_battery_storage`
`solar_wo_integrated_battery_storage`
|
| | `unknown_energy_storage` |
| `wind` | `wind_w_integrated_battery_storage`
`wind_wo_integrated_battery_storage`
|
### Inconsistent interchange
Balancing authorities do not always report symmetric, internally consistent interchange.
For example, entity A might report receiving 100 MWh of energy from entity B, while
entity B only reports sending 50 MWh of energy to entity A, so the data as reported do
not satisfy conservation of energy, and may need additional processing for use as
energy system modeling constraints.
### Localization of timestamps
In the raw EIA-930 data both UTC and local timestamps are provided. We have retained
only the UTC times, because the timezones used by the balancing authorities do not
always correspond to their physical location, and the localized times include gaps and
duplicate hours due to daylight savings time shifts. The reporting timezone used by each
BA can be found in the [core_eia_\_codes_balancing_authorities](../data_dictionaries/pudl_db.md#core-eia-codes-balancing-authorities) table, and can be
used to re-localize the UTC timestamps if needed.
### Inconsistent totals
See PR [#3584](https://github.com/catalyst-cooperative/pudl/pull/3584) and [this comment](https://github.com/catalyst-cooperative/pudl/pull/3584#issuecomment-2080346566)
for an exploration of the following irregularities:
* In some cases the sum of reported net generation by energy source does not equal the
total reported net generation for a BA. In most cases it’s close, but for some BAs
(typically the smaller ones) it can be off by factor of 10 or more.
* The sum of reported demand by subregion is very close to being equal to the total
demand for a BA, but the correspondence isn’t perfect.
* The sum of reported interchange with adjacent BAs does not always equal the total net
interchange for a BA.
* Directly calculating demand as the total net generation minus interchange sometimes
results in very different values than the reported demand.
### BA consolidations and retirements
Many smaller balancing authorities have been consolidated into larger organizations over
the years, especially into the Southwest Power Pool and MISO. As a result there are a
number of historical BA codes that **also** appear in the BA subregion codes. Compare
these tables:
* [core_eia_\_codes_balancing_authorities](../data_dictionaries/pudl_db.md#core-eia-codes-balancing-authorities)
* [core_eia_\_codes_balancing_authority_subregions](../data_dictionaries/pudl_db.md#core-eia-codes-balancing-authority-subregions)
### Pseudo-ties and dynamic scheduling
Form EIA-930 data attempts to represent, as purely as possible, a physical picture of
supply, demand, and energy flow (interchange) on the U.S. electric grid. For the
purposes of Form EIA-930, interchange is the physical flow of electricity metered at the
tie line boundaries between electric systems. Net generation is a metered value of the
output of generating units in the BA. Demand (or load) is a calculated value based on
the activity occurring within the tie line boundaries of the electric system.
This physical picture of BA operations is not the way many electric systems account for
and report how they dispatch their generators, manage interchange, and balance their
systems. Utilities sometimes enter into commercial arrangements with other systems
including:
* **Dynamic scheduling:** the dispatch of generating units located in one system to
serve customers in another systems
* **Pseudo-ties:** the transfer of electric energy through virtual interchange, where no
direct physical connections exist between the two systems.
If operating data were reported with these commercial arrangements incorporated, then:
* Demand might include demand served outside the BA’s tie line boundaries or exclude
internal demand served by other BAs
* Net generation might include the output of generating units outside the BA’s tie line
boundaries or might exclude the output of internal units dispatched by other systems.
* Energy flows might be reported between BAs that are not directly interconnected.
### BA Codes and Names
Note that the BA codes that appear in the EIA-930 are based on official NERC BA codes
and so may not match the current full name of BAs because of changes in corporate
ownership, mergers, or other business transactions.
### Other EIA-930 Projects
* See the [Grid Emissions project](https://github.com/jdechalendar/gridemissions/)
from [@jdechalendar](https://github.com/sponsors/jdechalendar) at Stanford for one example of managing irregularities in
the reported data, and the related paper [A Physics-informed data reconciliation
framework for real-time electricity and emissions tracking](https://arxiv.org/abs/2103.05663) for a method of reconciling the reported data
with physical constraints.
* The [Open Grid Emissions Initiative](https://github.com/singularity-energy/open-grid-emissions)
from [@grgmiller](https://github.com/sponsors/grgmiller) at [Singularity Energy](https://singularity.energy) builds
on the above method. See the [OGE Documentation](https://docs.singularity.energy/docs/open-grid-emissions/about_ogei-about-the-open-grid-emissions-initiative).
## PUDL Data Transformations
To see the transformations applied to the data in each table, you can read the
docstrings for [`pudl.transform.eia930`](../autoapi/pudl/transform/eia930/index.md#module-pudl.transform.eia930) created for each table’s
respective transform function.