Where does the electricity go?
Part of the own consumption stays unexplained because not everything is measured.
Concern 1 · own consumption
Who uses how much electricity at your plant?
Not every consumer has its own meter. The machine learning system splits the total consumption across the individual consumers and shows openly what is measured and what is estimated.
Sound familiar?
Three questions operators ask.
What is measured, what is estimated?
A number without origin does not help with decisions.
Where does a meter pay off?
Whoever knows where the gap sits can retrofit selectively.
How the learning system answers
Answers with the origin of the number.
For every share the system says whether it is measured, extrapolated, modelled, carried forward or unknown.
Which consumers used the most electricity last month?
Consumers A and B are ahead. A quarter of the consumption is not recorded; the share is shown as “unknown”.
Source: meter series and balance, calculation in the report
Example with invented data
Questions you can ask
- What is each consumer's share?
- Which share is only estimated?
- How does the picture change over the year?
- Where is a measuring point missing?
Every number comes from a fixed calculation, not from the language model. Every answer names its source.
Even without a meter at every consumer
The system splits it up and says openly what is measured and what is estimated.
What it looks like
Shares with evidence class.
The colour strength shows how well a share is evidenced. The unrecorded remainder stays visible; it is not hidden.
Who uses how much?00:00
Load now–
Recorded–
Not recorded–
Example with invented values
Every measuring point counts
Whoever knows where the gap sits can retrofit selectively.
How it works
From measurement series to the split.
01
Connect measurement series
Meters, control system and sub-meters deliver the series.
02
Run the fixed calculation
A fixed calculation splits and estimates only where nothing is measured. The calculation path can be inspected.
03
Answer with source
The language model phrases the answer and names the calculation as the source.
Learning system and Pseudomizer
Two building blocks, one goal: answers without handing out originals.
The learning system (RAG)
A system of your own with its own storage. It reads your documents and measurement data, answers in plain sentences and names the source. At night it asks itself new questions and reports gaps.
More on the local AIThe Pseudomizer
If you also want to use a large cloud AI, the Pseudomizer creates a shadow copy and checks it with subject-specific check chains. The original stays with you. Leak rate at the exit: 0.00 %. Guards detect, the judge stage blocks when in doubt, healing cleans up.
To the PseudomizerMore concerns
What operators also want to know.
What to feed, and when?
Feeding recommendation next to the electricity prices: substrates, amounts and times.
To the pageWhen should the CHP run?
Electricity price outlook as a heat map, to plan operation against the price.
To the pageMay this go to the cloud?
Original versus mirror: use cloud AI without originals leaving the house.
To the pageWhere does the heat go?
Heat network at a glance: flow, return, output, storage. One click opens the chart.
To the pageBring us your biggest annoyance.
One conversation, one real task from your daily work. We use it to show what AI and automation can do for you.