How to Reduce Peak Demand Charges
A ladder of measures from free to capital-intensive, with the diagnostic that tells you which rung your site is actually on and the arithmetic for deciding between them.
Reducing a demand charge means changing the shape of a load curve, not its area. That is a different problem from energy efficiency, and it is often cheaper, because shifting a load costs less than eliminating it. This section covers the practical levers — sequencing and staggered startup, thermal storage, batteries, on-site generation, curtailment agreements, control settings and interval-data analysis — with the arithmetic for deciding which of them a site actually needs. It also covers the ones that fail: measures that cut kilowatt-hours while leaving the peak untouched, and control strategies that shave one meter while pushing the peak onto another. Every strategy here is judged against the same test, which is whether the next twelve bills fall.
The piece that carries the subject. Read this one first.
A ladder of measures from free to capital-intensive, with the diagnostic that tells you which rung your site is actually on and the arithmetic for deciding between them.
Everything underneath the pillar, in this subject area.
A demand controller predicts the interval average and sheds load before the target is passed. The concept is sound; the implementations fail in a small number of predictable ways.
You are not selling electricity. You are selling a commitment to reduce load on request, and the penalty structure for failing to deliver is the part worth reading twice.
Efficiency reduces the area under the load curve. Shifting changes its shape. They save money on different lines of the bill, and confusing them wrecks business cases.
Buildings store heat. Running the cooling plant harder before the expensive window and coasting through it turns that thermal mass into free storage.
The highest interval of the month is frequently the moment after a break, when everything restarts together. Spreading those starts over twenty minutes usually costs nothing.
Make cooling at night, use it during the peak. The cooling load is unchanged; the electrical load that produces it moves to hours where capacity is not being priced.
Solar cuts kilowatt-hours reliably. It cuts the monthly peak only when the sun cooperates in every interval that matters, and Berkeley Lab's modeling shows how rarely that happens.