To call or not to call? Optimizing DR dispatch strategies through data analytics

To call or not to call? Optimizing DR dispatch strategies through data analytics


By: SGN Staff


By Brad Davids


Warmer temperatures are making their way into weather forecasts across the country - heralding the start of baseball season and, right behind it, the summer peak season for electricity providers. Utilities and grid operators are once again gearing up their peak-shaving demand response (DR) programs. DR program managers need to balance the need for cost-effective dispatch with the need to have limited-hour DR resources available when needed - all while keeping customer satisfaction in mind. As a result, determining the right dispatch strategy before the heat of summer is a critical challenge for these programs.


In many cases, DR programs are designed to address system reliability and are only called when there’s an impending grid emergency. For these programs, the dispatch strategy takes care of itself, since specified emergency conditions trigger dispatch. However, for economic and peak-shaving DR programs, program managers must decide when to dispatch their DR resources such that events will align with system peak times and deliver the most value from the program.


Further complicating this situation, these DR programs usually have a fixed number of hours or days available per season. This is where best-fit "call or don’t call” criteria for a given program come into play. On the one hand, utilities need to avoid exhausting the DR resource by calling it too often. On the other hand, utilities want to be prepared with callable hours to use should a late-summer heat wave sweep through their service territory. Between limited hours, customer satisfaction, resource cost-effectiveness and customer fatigue, utilities have a lot of competing priorities when developing a dispatch strategy.


The key to success is taking a data-driven approach to determine the right dispatch triggers, striking a careful balance between program parameters, and planning ahead to make adjustments along the way. Two of my EnerNOC colleagues, Craig Williamson and Dana Haffner, recently shared their experience and thoughts in a webinar on the quantitative approaches they have taken to develop effective DR dispatch strategies. The EnerNOC Utility Solutions team implemented the approaches discussed in the webinar as part of two separate utility programs, each yielding successful results in balancing program objectives with customer needs.