Captive Solar Against a Three-Shift Load

How to read this sheet

The whole point is that generation and load are matched hour by hour rather than summed over a year. A unit is worth the tariff only if the mill is drawing at least that much at the moment it is made; every other unit is worth the export price, and on most schemes that is a third of the tariff or less. Sizing by annual consumption divided by specific yield ignores this entirely, which is why the proposals that arrive are usually built on it. The sheet also reports what the maximum demand does, because it is the question a mill asks second and the answer disappoints: solar displaces energy and the demand charge follows the peak, which on three shifts is at night. Expect the kilowatt-hour line to fall a great deal and the kilovolt-ampere line to hardly move. A typical-day model understates export. A real year is bright days when the plant makes far more than average and dull days when it makes almost nothing, and export is a convex function of output, so averaging the weather across the profile hides some of the surplus. Where the export price is far below the tariff the true benefit is therefore a little lower than reported here, and an hourly year of real irradiance is the way to settle it before signing. Not modelled: banking and settlement rules, which in several states let surplus be carried across months and change the export value completely; open access and wheeling charges; battery storage, which is the direct answer to the mismatch this sheet measures; curtailment when the utility refuses export; and the accelerated depreciation or capital subsidy that often decides these projects on tax grounds rather than energy ones.