Continuous energy analysis for commercial buildings — engineering-grade savings you can defend.
Your first building is analysed free: a screening-level report, one building per organisation, from whatever data you have. Interval or BAS data goes through Energy Agent; a year of bills goes through Utility Bill Analyzer. No commitment.
$43,500 a year, in one ordinary school. Not one of them tripped an alarm.
A year of one school’s electric meter, read in seconds. Engineering rules and machine-learning models find it; an engineer stands behind it. Watch these faults happen in the 3D school →
A synthetic school, simulated by the Energy Agent demo engine. Not a customer result.
The same school in 3D, from before dawn to 2 am: watch the load follow the day, lift the roof, and switch on the faults the scan found. There is a campus plant and an air handler too, and three games to play in them.
Watch this school waste $25,000 a year without a single alarm. Open the 3D model →
Also in 3D: a campus chiller and steam plant → · an air handler and its VAV boxes →
Play: Rogue Shift · Fault Detective · Lights Out
Compare against 6,436 buildings in EIA's CBECS 2018 survey. Free, runs in your browser.
Most of what we find costs little or nothing to fix. It just has to be found.
You send us the data your building already produces. We return a ranked list of what is wasting money, what each item is worth per year, and what to do about it.
96 automated fault rules read your meter and BAS data — air handlers, VAV boxes, chillers, RTUs, heat pumps, zones and the meter itself, including the Guideline 36 fault conditions FC01–FC16.
Every fault is priced against your actual utility tariff — including demand charges — with the method and confidence stated on its face.
Each number opens into its own calculation, with the assumptions listed. Your engineer can check the arithmetic; finance can trust the total.
A baseline model learns how the building uses energy from its own meter and the weather. Once a fix is made, the model says what the building would have used and the meter says what it did. The gap is the saving, reported with its uncertainty.
15,604 kWh verified over 100 days. 95% confident: 14,627 to 16,581 kWh. The saving built into this example was 15,620 kWh.
metered usewhat the model says it would have usedthe saving
Both run the same engineering analysis. One starts from your meter and BAS, the other from your bills.
Find out what is driving your energy use
Pelorus connects your interval meter data, building automation trends and weather, then runs an engineering analysis that pinpoints schedules, setpoints and equipment faults behind the number you just saw. You get a ranked action plan with savings and payback.
Best when a building is underperforming and you want to know why.
Turn a stack of utility bills into a portfolio view
Send us your bills. Pelorus extracts every meter, tariff and charge, audits rates for errors, and benchmarks each site so you can see which buildings deserve attention first, without installing anything.
Best for portfolios, or when bills are the data you have. The first building’s screening report is free here too.
Free tools and explainers, no sign-in
CBECS Explorer ranks your building against thousands of U.S. peers. The psychrometric chart walks through nine HVAC processes. The fault library explains all 96 rules, and the readiness quiz says which of them your data can run.
Built for engineers, operators and students.
Three minutes with the readiness quiz tells you how many of the 96 fault rules can run on the points you already trend, and which single point unlocks the most.
12 months of interval meter data. A BAS trend export too, if you have one — that is where the equipment faults live.
Typically a few days. Weather-normalised, priced at your tariff, benchmarked against comparable buildings.
A prioritised work list with owners and due dates, plus a report you can hand to finance without rewriting it.
Your team keeps doing its job. We run the analysis and bring you the findings.
Pick the one that bothers you most and send what you have: twelve months of interval data, a BAS trend export, or just the bills. You get a screening-level report of what is happening inside it and what it is worth. One building per organisation, no commitment to go further.
