How Stromfee Brings AI Into a Swimming Pool

A hotel pool, whirlpool and spa run around the clock, and much of that energy is spent when no guest is in the water. Stromfee introduces AI to these systems along a fixed path — inquiry, metering, plausibility check, mobile measurement, digitisation, a knowledge base and finally local control — so the changes are grounded in measured data rather than assumptions. This page describes that path for pools and hotels in Asia. It is general information, not legal advice.

Why pools and hotels in Asia

A wet-leisure area is an always-on load. A whirlpool held at 40°C runs 24/7, and its jets often keep running when nobody is sitting in it — in Stromfee's own example, that difference is described as paying roughly 180€ a day for a pool that could cost about 90€ a day at the same comfort level. Pumps, heating, jets, steam, lighting and sauna all draw power whether or not the area is in use.

In Asian grids the price of that power moves sharply through the day. As Stromfee's Asia program puts it, solar floods the grid at noon and vanishes by evening, so prices can swing from negative to triple-digit within hours. Stromfee describes its AI as the intelligence layer that reads those swings and turns them into value through forecasting, dispatch and battery optimisation — an approach it says was proven on a German fleet and is now offered for Asia. A pool is a good place to apply that layer because its loads are large, continuous and often shiftable.

Step 1–3: Inquiry, metering and plausibility check

The path starts with an inquiry. Before any control logic is proposed, Stromfee's stated principle is to work with the energy data the site already uses and to process it in real time, rather than replacing the existing infrastructure. That keeps the operator in control of their energy flows and keeps the first step light.

Metering comes next: the actual consumption of pumps, heat generation, jets and lighting is recorded so that decisions rest on measured values. The plausibility check then tests those readings for consistency — confirming that what the meters report matches how the pool is actually operated — before the numbers are used to justify any change. This ordering is deliberate: no optimisation is claimed until the underlying measurements have been checked.

Step 4–5: Mobile measurement and digitisation

Where fixed meters do not cover a component, a mobile measurement is taken on site to fill the gap — for example on a pump or a heating circuit that is not yet instrumented. This gives a temporary but real reading instead of an estimate.

The installation is then digitised: photographs and documentation of the plant room, pumps, heat pump and controls are captured so the physical system has a data twin. Digitisation turns a walk-through into a structured record that later steps — the knowledge base and the control logic — can rely on. Nothing about a guest is required for this; the subject is the equipment.

Step 6–7: Knowledge base (RAG) and local control

The collected data — meter readings, mobile measurements, photos and documentation — is organised into a retrieval-based knowledge base (RAG) specific to that site. This lets the AI answer questions and make recommendations grounded in the pool's own measured reality rather than generic assumptions.

The final step is local: the intelligence runs at the site so it can act on real-time energy data on the premises, keeping the operator in full control of their energy flows. Stromfee describes this integration as straightforward and quick to implement precisely because it builds on data the operator already has, rather than on a rip-and-replace of the plant.

What the AI actually does for a pool

Once the path is complete, Stromfee lists concrete actions for pool and spa. Pump optimisation: the AI analyses bather patterns and a variable-frequency drive adjusts pump speed automatically, which Stromfee states can cut pump electricity by 55%. Heating management: the AI forecasts heat demand from weather and booking data and runs the heat pump during low-price hours, stated as up to 60% lower heating costs. Presence control: jets, steam and lighting activate only when guests are actually using the wellness area, stated at 45% lower attraction costs. A sauna schedule builds an optimal infusion plan to shorten preheating.

These figures come from Stromfee's own pool-and-spa material and describe expected reductions from these measures, not guaranteed results for every site. They are also consistent in kind with what Stromfee reports for the wider hotel market — for instance, competitor Verdant (Copeland) is cited as reducing HVAC runtime by 45% during vacancy for about 18% total savings — but the pool figures above are specific to pool, whirlpool and sauna loads.

Privacy and scope

Presence-based control raises a data-protection question, so the design goal is anonymous operation: the system needs to know that the wellness area is in use, not who is using it. Occupancy is treated as a signal to switch jets, steam and lighting, not as a way to identify individual guests. The digitisation step documents equipment, not people.

This page is grounded in Stromfee's published GDPR AI guide for pools, hotels and wellness and is provided as general, anonymous information. It is not legal advice; operators should confirm the specific data-protection requirements that apply to their site and jurisdiction before deploying presence sensing.