How Graph Neural Networks Unlock the Future of Forecasting on The Electric Power Grid

The tech giant’s graph neural network is systematically benchmarked and up to 15% more accurate than traditional methods. Weather forecasts are the foundational layer for any power market participant. They are the basis for any load, solar and wind forecast, which themselves are a key component in price formation. On any horizon, they drive billion-dollar investment and operational decisions every day. Most weather forecasts today are derived from numerical weather prediction

With real-time prices falling, day-ahead participation rates for solar assets have reached 30%, with an average $1.50 per MWh improvement in returns.

How a developer/owner/operator of large renewable generation in California and Texas uses S&P Global Energy CERA Power Source Bidding Benchmarks to reduce evaluation time, improve their P&L, and support their trading operations. Bidding Benchmarks shows load, solar and wind profiles how different trading strategies might select different markets in the day-ahead market. Intro + Company Snapshot A leading player in the clean-energy sector boasts a substantial portfolio of over

How bigger data and bigger models deliver better results A large hydro-electric plant benefits from economies of scale. So do machine learning models. The prices to which generators and loads are exposed in wholesale power markets are driven by the marginal cost of producing and moving the power to where and when it’s needed, subject to constraints on the grid. The process, security-constrained economic dispatch, is run sub-hourly by independent system operators to determine

Backtested strategies and how traders can use them How will an incoming windy, cloudy cold front raise net load? How will a line outage affect congestion at your gas plant? Predicting how grid conditions and weather events will affect power prices is challenging, yet the often‑stressful decisions that power traders make nationwide are precisely what keep the grid running every second of every day. We wrote last week about how most decisions that a trader makes are based on pr

When predictions are perfect, the decisions are trivial. It follows that any successful optimization will depend largely on the accuracy of the predictions being used.

How graph-based models and transfer learning can improve price forecasts at new nodes with limited or no historical data The sun rises on a new wind farm in Texas The power grid is constantly evolving. In 2025, there will be 870 new assets that begin selling power to the grid, around 85% of which will be operated in a deregulated market with an associated price node. Not all may be new, but many will be. For developers and traders, forecasting prices at these new locations i

How forecasting algorithms can include upcoming transmission outages to better predict the congestion component of locational marginal price Substations make up a large part of the power grid and can cause significant changes in the solved optimal power flow when they go down for maintenance. Introduction to Congestion Locational Marginal Prices (LMPs) in energy markets are comprised of three components: energy (marginal system cost of producing one more unit of power), loss

Profit matters most when evaluating forecasting models The most important part of forecasting is how it improves downstream decision...

How historical forecasts can support developers and independent power producers to evaluate different trading strategies for their merchant assets Programmatic trading in power markets has two layers: the prediction layer and decision layer. The first layer is where forecast companies operate – training machine learning models to generate forecasts for what may happen in economic dispatch the following hours and days. Operators use these predictions to make decisions – wheth

A simple overview of the power grid, generation technologies and economic dispatch. The largest machine in the world is the electric power grid to which your lights, appliances, and soon your car are connected. It lives in the wires behind your drywall, atop the distribution lines that frame your street, through a substation and along tall picturesque transmission lines to power plants spanning the continent. It’s an infrastructure project that was built to bring power to you

The evolution of independent system operators in North America The market for electricity for much of the twentieth century was dominated by geographic monopolies, with consumers having no choice over which utility supplied their demand for power. In the 1980’s, some economists argued that this structure removed the incentive to operate efficiently and encouraged unnecessary investments, suggesting instead that the overall economy would be better off if the supply of electric