Engineering storage for weak grids: what nameplate capacity hides
Sizing batteries to load profiles rather than generation capacity is the single highest-leverage decision in distributed power design.
T. Adeyemi
Chief Operating Officer, Infrastructure
Nameplate capacity is the most quoted and least useful number in distributed power. It describes what an asset can produce under ideal conditions, which is a description of a laboratory rather than a substation in Ogun State in August.
What determines whether an industrial offtaker experiences reliable power is the relationship between the load profile and the storage configuration. A site with a sharp morning ramp and a sustained afternoon plateau requires a fundamentally different battery specification from one with an even draw across a three-shift operation — even where the two sites have identical annual consumption.
Our standard practice is to meter the site for a minimum of one full production cycle before finalising any storage specification. This frequently produces an uncomfortable conversation, because the metered profile often contradicts the offtaker's own description of their consumption. It also routinely changes the specification by a material margin in both directions.
The commercial consequence is that we occasionally recommend a smaller system than the client expected to buy. This is not a concession. Oversized storage in a weak-grid environment degrades on a cycle profile it was never designed for, and the availability failure arrives in year four rather than year one — by which time the specification decision has been forgotten and the technology gets blamed.
Availability, not capacity, is what the offtaker actually purchases. Engineering the asset around that fact is the whole discipline.