EPRI’s concept paper on the Integrated Grid outlined as one of four action plans the need for Interconnection Rules that preserve voltage support and grid management. But determining exactly what is needed under these new rules is both challenging and under considerable discussion and debate.
For example, ride-through is a hot topic these days in discussions of distributed energy resources (DER), especially in standards communities, such as IEEE. Having been well-addressed at the bulk level (by NERC, IEEE, and others) and for critical loads (such as industrial equipment), new technologies are being enlisted to sustain operation during abnormal voltage or frequency. In some ways, discussions about these new technologies applied at the distribution level are even more complicated than discussions about the bulk system, due to a broad array of resources, interfaces with the power system, and competing expectations of system engineers. Often, these discussions leave more questions than answers.
For DER that are defined and commissioned under the prevailing standard in the United States (IEEE-1547), expecting ride-through is a huge logical departure from the existing “must-disconnect” requirements. Now, DER units must have a three-tiered structure where some operating areas are “must remain,” whereas others are “may disconnect,” and yet others are “must disconnect.” Fulfilling these requirements is complicated yet achievable. However, ambiguity in the eventual ride-through requirements would complicate efforts to design DER equipment to comply with the requirements. Standards should clearly state DER requirements, even if they are more complex than in the past.
In that vein, the standards community must face two big questions before robust discussions can really move toward a conclusion regarding ride-through of DER equipment:
What do we mean by ride-through?
The answers to this question varies, but the ride-through behavior of DER equipment during an abnormal condition at the point of connection to the grid generally falls into one of three categories:
- The DER stays connected but is not required to provide power. This is one of the safest options and is simple to understand and define in a standards document. This also prevents DER from unnecessarily feeding a fault, and it is ready to provide energy again after the abnormal condition has been remedied.
- The DER stays connected and continues to operate in an “as-you-were” fashion. This is more challenging to implement than the previous option. Two primary concerns are the potential for feeding a fault, limiting inverter current during a voltage sag, and adjusting to the behavior of an “as-available” resource.
- The DER modifies real and reactive power to counter the abnormality (it provides reactive power to counteract a voltage sag or reduces real power during an over-frequency event). This is the most complex form of ride-through, combining what has been referred to a “smart inverter” functions (such as volt-VAR or frequency-watt) with a requirement to remain connected and supply power. This behavior is much more complicated to design but enables DER to be more autonomous.
There are conceivable operating conditions where each of the above responses would be optimal. How do we translate a set of possible scenarios into a uniform recommendation for a standard? For instance, on a distribution feeder with a small amount of DER installed, the best option may be simply for the DER to stay connected but de-energized during a distribution fault. However, if the situation is an under-frequency event on a system with a large penetration of DER, the system would benefit if these units continued to provide real power. What are the proper signals for ride-through behavior? How does a DER distinguish between different system events? How do we balance the requirements for both distribution and bulk systems?
How should ride-through expectations be set?
Even with a sufficient definition of ride-through, designers of DER equipment will need to know: Which types and sizes of DER should be responsible for providing ride-through? To what degree should they provide ride-through? Based on what criteria? There is tension between remaining technology agnostic with ride-through requirements and the ability to maximize potential contribution of each resource to system performance.
If defining a uniform expectation for all DER, IEEE 1547 may have to be more conservative (less ride-through) than an inverter-only standard (the proposed California Rule 21). However, synchronous DER may have additional headroom to provide frequency response. This is due to both the characteristics of the source, such as solar or wind, and the capabilities of the interface between the source and the grid.
A potential compromise may be to define a generic set of resource “types,” each with an expected ride-through characteristic. Specifying the required types at different power levels could offer a blend of ride-through expectations without tying requirements to any one resource.
As part of the Integrated Grid Initiative, EPRI is developing a series of whitepapers on recommendations for interconnection requirements, including ride-through.
What do you think? How should ride-through be defined? Is there a good way to move from isolated scenarios to a unified standard? Should solar inverters be included under the same standards as synchronous DER or storage systems?