Again, the claim on the website is "has the potential to power over 91,000 homes". "up to", "has the potential", etc. But let's do some quick back of the envelope maths.

91,000 homes would be about 91,000 * 3,000 kWh / year = 273 TWh / year. You need an average of 31 MW over a year to produce this, which roughly tracks 350 MW with ~10% capacity factor. So over a year this seems to be correct.

On a day to day basis however the project only includes 700 MWh of storage. That's about 24h of the electricity consumption of 91,000 homes. A single cloudy winter day and those batteries are empty. So if you look at shorter time periods than a full year there's a huge variation in how many households can actually be reliably powered by such an installation. It's likely somewhere between 1,000 (several days of cloudy winter) and 350,000 homes (sunny summer).