Nobody is relying on 1000 kW from that array. If it is a 3x overbuild, then they are counting on only 300 kW from it, with any extra going to charge storage or put on a transmission line to sell. Their shortfall under stormclouds is then 300-75=225 kW. After their storage has been depleted, if power from elsewhere is already spoken for, that is how much they will draw from their backup generator. At night, when the array produces zero, demand is lower, but anyway their backup generator can take the whole 300 kW load when necessary.
But it is not often necessary. Often they can fill the shortfall from storage or from a transmission line with power from elsewhere, maybe wind, or maybe even from the next town's storage if they predict they won't need it right away. The town will deliver excess stored energy they have on the same terms.
After developing some experience, and prices for panels have plummeted further, maybe the town builds out another 2x. Then they generate 125 kW under storm conditions, and the shortfall is only 175 kW. The rest of the time they can charge their storage faster and then sell more to the grid.
How much to spend overbuilding generation vs. how much on growing storage capacity is a complicated calculation. Both are capex for equipment that cuts hard-to-predict costs. Building more later gets you lower prices, but only starts saving you money after it's built.
Of course in practice all the solar and wind farms push to the grid at the bid price, and the town draws power from the grid at the asking price. Storage and gas generators push power to the grid when bid price exceeds their profit threshold. Storage draws from the grid to charge when the asking price is low enough, which is never when the gas generator is pushing. Nukes never see the bid price exceed their cost, but must take what is offered or shut down; all not heavily subsidized via tax will shut down.
Anything you write that is not silly, I promise not to call silly.