Skip to main content

Power Model Integrator Reduces Forecasting Errors


Grid coordinators have the daily challenge of forecasting the need for and scheduling exchanges of power to and from a number of neighboring entities. The sum of these future transactions, called the net interchange schedule, is submitted and committed to in advance. To set their schedules, forecasters rely on a combination of personal experience, historical data, and often a preferred forecasting model, which often results in unsatisfactory forecasts and leads to inefficient market operation.  Adding to this forecasting challenge are affects from weather, renewable energy fluctuations, and variability in power exchanges between neighboring organizations at any given time.


Researchers at Pacific Northwest National Laboratory developed a software tool that greatly improves the accuracy of net interchange predictions.  The tool, called the Power Model Integrator, adaptively combines the strengths of different forecasting models continuously and in real time. This addresses a variety of scenarios that impact electricity use, from peak periods during the day to seasonal swings. Performance of the tool was tested against five commonly used forecasting models processing a year's worth of historical power system data.  For forecasts one-to-four hours out, they achieved a 30-55 percent reduction in error.  And with longer-term forecasts—the most difficult to accurately make—the tool also performed best.


With little time for corrective actions, incorrect forecasts can result in millions of dollars in wasted energy costs and affect the stability of regional power markets. The new forecasting tool delivers up to a 55 percent increase in accuracy, allowing utilities to operate closer to the system limit, while minimizing penalties for violating operating standards. The tool completed production testing and is moving into commercial use at the Independent System Operator level, helping improve grid reliability and efficiency.

PPower Model Integrator software tool

Quick Facts

Date: August 1, 2016
Type of News: Project Highlight