There is a growing community of engineers and analysts that use production cost models (grid simulation tools) to analyze future power systems. We at NREL do a lot of grid simulations looking out over the next several decades and often need to implement technologies or grid services that aren’t represented in the “out of the box” software packages. We often have to create our own models and constraints, and we just published a pair of reports that provide detailed mathematical formulation on how we have implemented two grid technologies. While we use the PLEXOS production cost model, we hope that our methods are transparent enough that they can be represented in other unit commitment models, including commercial packages.

The first discusses implementation of concentrating solar power with thermal energy storage (CSP/TES). While we have been publishing analysis of CSP/TES using our formulation for a long time, we had never gotten around to providing explicit details of how we do it, so this publication corrects that. The second publication discusses how we have implemented primary frequency response. This might be of growing interest as we move to more inverter-based generation, where provision of frequency responsive reserves may be a binding constraint in grid operations. Ultimately our goal is to implement a physics-based fast frequency response constraint to understand the impact of inverter-based machines on grid economics. But that’s for another post.