BRENNEN

Industries / Energy, water and waste / Electric transmission and distribution

Smart metering

Smart metering is the business of fitting electricity, gas and water meters that record use and send it to the utility on their own, and of the networks, data systems and billing that turn those readings into bills, prices, outage maps and leak and theft alerts.

1965

Transitel International reported a sixteen-month test of its Automatic Meter Reading System, which read household utility meters over telephone lines in Owosso, Michigan, and found it "technically feasible" and compatible with telephone company equipment (Brothman, Reiser, Kahn, Rittenhouse and Wells, IEEE Transactions on Communication Technology, 1965).

manuscript.isc.ac

28

Milestones achieved across 4 domains.

7

To watch for: announced, not yet confirmed outside.

6

Open deliverables. Nobody has shipped them.

In one minute

Where it started

For most of the 20th century a meter reader walked to every meter and wrote down the dial. In the 1960s and 1970s engineers read meters over telephone lines in trials, and in 1986 Itron shipped a small radio for gas meters that let readers collect figures by walking or driving past. Power-line signalling and, from the 2000s, two-way meter networks followed, so the utility could read every meter every day or every few minutes.

Where it stands

IoT Analytics counted over 1.06 billion smart electricity, gas and water meters worldwide at the end of 2023, about 43% of electricity meters; Counterpoint Research put the installed base at more than 1.8 billion at the end of 2024. The two firms count differently and their figures do not agree. The US Energy Information Administration counted 147.1 million advanced meters out of 169.8 million US electricity meters in 2025, about 87%. The data now sets hourly and half-hourly prices, reports outages, finds leaks and flags theft, while failures in Saskatchewan, Britain and the Netherlands showed what goes wrong when meters, networks or bills are rushed.

What is open

Much of the published work on meter data cannot be checked. Theft detectors are scored on a public Chinese dataset in ways that leak test answers, no public household dataset of water leaks with confirmed labels exists, and no open benchmark tests how well data systems catch and fix bad readings. Public smart meter datasets, open time-series models and cheap radio receivers let a small team build these tests now.

Domains

AchievedWatch forOpen

Open in smart metering

3 of 6 open deliverables

Open · Meters

Open waveforms for meter testing

A public library of the distorted current waveforms that household appliances draw (dimmed LED lamps, chargers, heat pumps), with an open simulation of a Rogowski-coil meter and a Hall-sensor meter that predicts each meter type's reading error on each waveform.

Built with what exists

The artificial test waveforms defined in the 2021 paper; commercial plug-through measurement adapters and clamp-on current probes for new recordings at the socket, with no wiring work; Python with NumPy and SciPy; Zenodo for publication.

Credit · unclaimed

Open · Communications

Measure meter radio range

A public, repeatable test of how often Wireless M-Bus meter messages (the 868 MHz radio format used by many European water, heat and gas meters) are received at each distance and through each kind of wall, for several common receivers.

Built with what exists

Meters the team owns; an RTL-SDR software radio running rtl-wmbus, and two USB receiver dongles; wmbusmeters; a laser rangefinder; Python. Receive only, on the licence-exempt band, and only the team's own meters.

Credit · unclaimed

Open · Data and billing

A test for bad-reading checks

A public answer key for "validation, editing and estimation" (VEE), the rules every meter data system runs to catch bad readings and fill gaps: real half-hourly readings with each fault (spikes, stuck values, clock shifts, meter swaps, runs of zeros) labelled by hand, plus a scoring script.

Built with what exists

The London Datastore smart meter data; Python with pandas; a labelling tool such as Label Studio; a rule-based VEE baseline written from published utility rules.

Credit · unclaimed

Where it sits