PJM Generation Mix by Fuel

93,886 intervals, 2016-01 to now. Columns: coal_mw, gas_mw, oil_mw, nuclear_mw, hydro_mw, wind_mw, solar_mw, storage_mw, multiple_fuels_mw, other_renewables_mw. Wide rather than long: a whole interval's mix is one row, so shares are computed across columns in that row and the timestamp does not repeat once per fuel. "Which fuel is largest" is greatest() across the columns, or simply reading the row.

Query contract

Select one exact event_time and read every fuel column from that row. A mix assembled from more than one timestamp is not a mix that ever existed.

Analysis rules

Compute each fuel's share against the sum of the fuel columns in that same row, not against demand and not against a total from another interval.For a change over a period, resolve complete buckets and compare their canonical observations; do not query two wall-clock instants separately.

Interpretation

PJM publishes all eight fuels the other markets use plus two more: multiple_fuels_mw (switchable units that can burn gas or oil, PJM's analogue of NYISO's dual_fuel_mw) and other_renewables_mw. Do not fold either into gas or into a residual — PJM reports them as their own categories deliberately.PJM publishes fuel_percentage_of_total per row, which implies a total, but does not publish the total itself. Sum the ten fuel columns for a total; do not divide a fuel's MW by its percentage, which recovers a total that disagrees with the column sum by rounding.

Download

Query it in SQL, download it, or read it with an API key. Download data

MCP

Connect an MCP client and load the contract with describe_data before calling query_data.PJM MCPAPI reference

HTTP endpoint

POST /api/v1/datasets/pjm_generation_mix/query

Full history

Delivered to Snowflake, Databricks or your own storage on a schedule by Data services.

More coverage

Request data coverage

Complete

100.0%

Gap days

0

Days with data

3,961