In todays world, using yesterdays forecasting technology may not be the right answer. In most of the world, forecasting is done top down. The data did not exist to do any other method of forecasting in the past. Monitoring sites like the CalISO and the IESO shows that while for a whole market top down forecasting still works, the differences between the forecast and the reality are growing, at least at the interval level. With the deployment of more sensors in the grid, from smart meters to transformer monitors to distribution substation SCADA, it is now possible to think about turning the forecasting techniques on their head. No longer restricted to top down forecasting, bottom up forecasting allows many advantages that top down does not.
Starting with the ability to see areas in the grid that are heavily loaded and require load control or other techniques to improve system security. Then there is the ability to do a better job at forecasting ancillary services, and peak load demand, allowing the localization (where possible) of the generation. Additionally as distributed generation sprouts on rooftops, bottom up forecasting will help determine how much it is helping or hurting the overall load, since many of the distributed resources will reside on only one phase of the system.
The Smart Energy Alliance Forecasting solution provides a way to do bottom up forecasting using realistic assumptions and tools. As your data sources become better, shouldn’t your information and knowledge get better too?
