A treatment plant’s electricity cost is set by a handful of decisions: when the pumps run, when the blowers turn down, when the battery charges or discharges, and whether the biogas engine runs or idles. The control algorithm makes those decisions against the market, the tariff, and the plant’s physical limits, every five minutes, without an operator watching the price screen.

Three price signals, one effective cost

Every plant sits under at least one, and usually several, price signals at once:

• The wholesale spot price: a new price every five minutes, set by supply and demand across the grid. (See The NEM Explained.)
• The retail tariff: time-of-use rates that vary by peak, shoulder, and off-peak periods. (See Electricity Tariffs Explained.)
• The maximum demand charge: a few peak intervals each month set the network charge for the whole billing period.

The algorithm reads whichever signals apply to the site and combines them into one effective cost. Which dominates depends on the site’s commercial arrangement: a plant on the spot market responds to the five minute price, one on a retail tariff responds to the time-of-use calendar, and both respect the demand charge.

The plant as a storage battery

Flexible load exists because treatment processes hold inventory. An inlet wet well and effluent storage reservoir hold flow that can be pumped now or later, within their operating ranges; sludge holding tanks can accumulate solids ahead of thickening or dewatering; and biogas storage banks gas for the engine. Each of those volumes is the plant’s energy storage.

Rather than sitting at one constant operating set point, the control system adjusts the set points continuously based on energy cost. Cheap electricity lowers the target level for a tank, so the pumps draw it down and process more while power is inexpensive; expensive electricity raises the target, so the tank stores the incoming flow. The same logic applies to each stored volume. The response is smooth, not an on-off switch, so the plant tilts toward cheap windows and away from expensive ones without breaching any asset’s limits.

Looking ahead, not just reacting

Reacting to the price now is not enough, because the plant has inertia: water takes time to move, a battery to charge, a biogas engine to warm up. The algorithm looks ahead, using forecasts of price, solar, weather, incoming flow, and the scheduled timing of product offtakes, and positions the plant in advance.

• An hour ahead, it holds the plant ready for the next high price interval and pre-empts storm inflows by drawing the tanks down.
• A day or more ahead, it positions the reservoirs, sludge holdings, and battery to ride out known peaks and to make best use of forecast solar and price.
• On a five minute basis, it dispatches pumps and battery against the live price.

Product offtake timing matters as much as price. A scheduled biosolids offtake or recycled water offtake is part of the forecast: the algorithm makes sure the relevant tank is at sufficient level to supply the scheduled offtake quantity, without letting it climb so high that upstream parts of the plant are forced to stop.

Where the price is never allowed to win

The commercial objective is always subordinate to the process. The algorithm carries the plant’s physical constraints at every step:

• Reservoir and wet well limits: hard low and high levels that are never breached, however extreme the price.
• Storm handling: when large inflows are coming, the tank is drained early so capacity is available, and flood risk is the highest priority signal of all.
• Hydraulics and pump behaviour: real pipework, friction, and pump curves set what flow is achievable, so the algorithm never asks for what the pipework cannot deliver.
• Pump wear: start and stop counts are damped so pennies on electricity are not spent on switchgear and impeller wear.
• A maximum import cap: where the site caps grid draw, load is shed to respect the cap without endangering the process.
• Operator override: operators retain full manual control at all times.

Every asset in one decision

Solar, battery, blowers, and cogeneration all join one coordinated decision against a single cost signal. Solar offsets grid import whenever the sun is shining. The battery charges when power is cheap or from solar surplus, and discharges at high prices or when a peak interval threatens. Blowers back off when electricity is expensive, while always holding the dissolved oxygen the biology needs, and make up the shortfall within the biological limit. And biogas generation runs when the price justifies it, banking the gas when it does not, with waste heat coordinated back to the digester. Every asset stays within its operating envelope.

Designed by simulation, not by guesswork

Before it runs live, the site is built into a simulation model, validated against actual plant data, and run over a full year of real prices, weather, and flows, so the settings are tuned for the whole billing period, not a single interval. That is the point of Integrated Industrial Energy Control: the logic that runs live is the logic that was validated in simulation, so a client receives a quantified forecast of the outcome, including the bill, the demand peaks, the pump starts, and the overflow risk, before committing to anything. The delivery process is set out in How We Work, and the commercial routes into it are described on Energy Management.

What this means for your site

A plant under price-responsive control does not need someone watching the market to run it profitably. The decisions are made automatically, within limits the process itself sets and enforces, and the expected outcome, including a lower bill, fewer peak intervals, and no lost effluent quality, is quantified in advance against your actual tariff and exposure. Whether your site is best served by wholesale market flexibility, tariff optimisation, or a new retail offer depends on its circumstances; the control method underneath is the same. You can see it in action on a full year replay, with a plant responding to real wholesale prices across four scenarios, on the sample simulation page.