Storage should be placed in at least a couple of categories: seasonal and short-term.
Sort-term storage I am trying to add to the model based on a report of a battery with 10MWh of storage and a 6 MW output - for modelling assumptions, I assume it can only charge at 50% of capacity - for estimating the capacity cost I have used 12MWh for the price of the 10MWh in this article. I haven't chosen to price fuel becuase I only charge the batteries during surplus baseload generation.
Seasonal storage is, in my opinion, no worthy of exploring in a long-term energy plan.
The most basic type of seasonal storage is a reservoir.
This need no be an articificial reservoir. My understanding is Quebec is challenging Newfoundland and Labrador's project on the Lower Churchill falls because Quebec claims full control of the Churchill Falls reservoir for not only seasonal storage, but to store power for future years! The conflict itself explains the rarity, and complexity, of seasonal storage.
There are other forms of seasonal storage: Denmark is, I believe, working with storing energy thermally for upcoming seasons, and compressed air in various forms is discussed as well. All these approaches, it seems to me, would be enormously expensive to provide capacity for relatively few peak hours annually.
Back in Ontario, Northland Power has a proposal to use an abandoned mine as an integral part of a pumped storage project. From the few details I know of the Marmora project, to integrate it in this calculator I'd reference it as 1600MW with a capability at peak of 25% (this would translate into running for 4 hours at 400MW, which is the claim); and the net revenue requirement I'll guess at $42000/MWyear. This is far more attractive than the battery preset - it's also site dependent and therefore not readily replicable - nor is likely accounting for the cost of connection to the broader grid.