Example system · Energy-intensive sites
Energy and peak control
Cut the peak, not the output.
Example system Energy-intensive sites
Cut the peak, not the output.
Sense. Sub-meters on every major load: compressors, chillers, furnaces, chargers. Alongside them, the tariff, the weather and the production plan.
Illustrative system. Simulated data, not client results.
Inside the engineering
The problem
Demand charges and time-of-day tariffs mean the same unit of energy can cost several times more at the wrong hour. Most sites can see the bill, but not which load caused the peak, or which loads could have moved.
Why it is hard
- 01Demand depends on weather, shift patterns and the order book at once.
- 02Some loads can move and some cannot, and the rules live in people’s heads.
- 03A plan that saves energy but delays production is worse than no plan.
- 04Control has to fail safe: when anything is unclear, the plant runs as normal.
What we build
From signal to decision.
- 01
Sense
Sub-metering on major loads, connections to building and process controls, and feeds for tariffs, weather and the production schedule.
- 02
Learn
A transformer forecaster for site demand at fifteen-minute resolution, with a model of each flexible load and the limits it must keep.
- 03
Predict
Tomorrow’s demand curve, the peaks that matter, and an optimised schedule for flexible loads within production and temperature limits.
- 04
Act
A daily plan for approval, automatic setpoint changes where you allow them, and a monthly account of what was saved and why.
Before we start
What needs to be in place.
- Interval data from your main meter, ideally every fifteen minutes.
- At least one load that can move: refrigeration, heating, charging or batch work.
- Access to controls, if loads are to be moved automatically.
How the project runs
From survey to live.
Fixed price once the specification is agreed. Every repository, model and account is yours from the first day.
- Survey1–2 weeksTariffs, bills, major loads, and which of them can move.
- Meter2–3 weeksSub-metering and data feeds, checked against the utility meter.
- Model4–6 weeksForecasts and constraints, with advisory plans checked against reality.
- Go live1–2 weeksDaily plans, automation where approved, and savings reports.