Skip to content
Technology transformation expertise firm · ARDIANE Group Innovate, secure, succeed

Industries

Digital and the energy transition meet in a network's measurements and what they require of the information system

In brief

Digital and the energy transition meet on four measurements of an energy network: metering, consumption and generation forecasting, balance at every moment and the condition of equipment. Management Square describes for each one how to read it, what it requires of the information system and what it leaves in the shadows.

Page contents

An energy network is managed through what it measures. The quantities delivered at each point, those expected for tomorrow, the gap between generation and consumption at every moment, the condition of equipment: it is on these quantities that digital technology and the energy transition meet. Article L. 100-1 of the French Energy Code indeed has energy policy contribute to a European energy Union “by means of the development of renewable energies, physical interconnections, flexibility resources of the electricity system”, among other things.

Each of these four measurements has a reading, a specific requirement of the information system and a blind spot. Metering requires collecting and protecting fine-grained series. Forecasting requires reliable histories and external data, including meteorology. Balancing requires flows processed without delay. Maintenance requires an accurate asset register. A fifth term remains, which the first four make grow: what digital technology itself consumes.

The articles of the Energy Code cited here were read on 13 November 2026 in the LEGI database of DILA’s open data (archive of 8 November 2026). The translations of quotations are ours.

Metering data, the raw material of the other measurements

Point 7 of Article L. 322-8 (version in force since 14 March 2026) entrusts the operator of an electricity distribution network with metering activities: supplying, installing, metrological checking and renewing meters, and “the management of the data”. Article L. 341-4 requires it to make available to consumers their metering data, alerts linked to their level of consumption and points of comparison. The operator also guarantees suppliers access to consumption data “downstream of the meter and in real time, subject to the consumer’s consent”. For gas, Article L. 453-7 provides for interoperable metering devices and comparable provision of data.

In our reading, these articles imply for the information system keeping time-stamped series for each delivery point, knowing for each supplier whether the consumer’s consent exists and on what date it was given, and producing, on the request provided for by the text, data “in anonymised form aggregated at building level”. When linked to an identifiable person, these series also fall under the General Data Protection Regulation, as our page on energy and utilities points out.

The metering figure has its limits. A missing reading may be replaced by an estimate: a series that does not distinguish measured from estimated values allows neither a forecast to be evaluated nor a dispute to be settled. The share of values actually read, by period and by point, is therefore tracked alongside the volumes, with the indicators presented in our article on data quality metrics. Finally, a delivered volume says nothing about the use that consumed it.

Forecasting, then balancing at every moment

Under Article L. 321-10, the operator of the public transmission network “ensures at all times the balance of electricity flows on the network”. It may modify the dispatch schedules taking into account the economic order of precedence between the balancing offers submitted to it. Over a longer horizon, Article L. 321-6 has it draw up a ten-year development plan based on assumptions about changes in generation, consumption and exchanges.

Two measurements answer each other. Consumption and generation forecasting is judged by its gap from actuals, according to the error and bias conventions detailed in our article on demand forecasting. Generation that depends on wind or sunshine brings weather forecasting into the input data, with its own uncertainty. The gap from balance, for its part, is measured continuously, and processing it falls under a streaming architecture whose cost and delay are discussed in our article on real-time data architecture.

Flexibility adds a delicate measurement. Article L. 341-4 calls for network usage tariffs that encourage customers to limit their consumption during the periods when consumption is highest at national level, and that may also target local peaks; Article L. 322-8 has the distribution operator contribute to monitoring demand response perimeters. Curtailed consumption did not take place: it is calculated as the difference from an estimated baseline consumption, and the result depends on the estimation method. A balance held at national level, for its part, says nothing about a local constraint: the distinction between national peak and local peak appears in the text itself.

The condition of equipment, measured to decide on maintenance

Article L. 322-8 also charges the distribution operator with operating its network and “ensuring its upkeep and maintenance”. The measurement here concerns equipment: signals transmitted by sensors and supervisory systems, intervention histories, age and conditions of use of each asset.

It requires of the information system an accurate asset register, in which each measurement and each intervention is linked to the right piece of equipment, and a data flow that goes up from industrial IT to the management platforms without opening a command path in the opposite direction. Our entry on IT OT convergence describes this convergence and the segmentation it requires; the security of entities falls under the NIS2 directive, which covers energy in particular, and connected products installed on the network may fall under the Cyber Resilience Act, which covers products with digital elements.

Two indicators can account for it: the share of equipment whose condition is known at a given date, and the time between an alert and the intervention that deals with it. The absence of an alert remains ambiguous. It may mean equipment in good condition as well as a silent sensor, hence the value of monitoring the reception of the signals itself. Bringing these sources together in a governed data platform, and deciding on the flows that cross the boundary between the two kinds of IT with an IT architecture approach, determines how these measurements can be used.

Digital and the energy transition, the account of digital itself

Collecting at fine intervals, keeping for a long time, recalculating often: each of the preceding measurements mobilises servers, storage and networks, whose manufacture and operation have their own footprint. This term cannot be deduced from any of the network’s measurements. It is measured separately, through an approach whose steps are described in our article on sustainable IT in business.

The two accounts nevertheless meet on two design choices. The measurement interval is set according to the decision it serves: a quarter-hourly series does not inform a monthly bill at a single price any better than a meter index, whereas it does inform demand response. The retention period is set in the same way, and the requirement to minimise personal data points in the same direction. Writing down, for each series, the decision it feeds and the period during which it remains useful gives a common basis to data quality, to the protection of subscribers and to the sobriety of the information system.

FAQ

Frequently asked questions

Another question? Ask us

Who manages electricity metering data?

Under point 7 of Article L. 322-8 of the French Energy Code, the distribution system operator carries out, within its service area, metering activities and manages the corresponding data. Article L. 341-4 requires it to make these data available to consumers and to guarantee suppliers access to consumption data, subject to the consumer's consent.

Why is demand response difficult to measure?

Demand response corresponds to consumption that did not take place. It is therefore calculated as the difference between an estimated baseline consumption and the observed consumption. The result depends on the method chosen to estimate this baseline, which must be written down and kept with each published figure.

Can digital technology serve the energy transition when it consumes energy itself?

The two are measured separately. A network's measurements inform operational decisions; the servers, storage and networks that produce them have their own footprint, linked to their manufacture as well as to their operation. Setting the measurement interval and the retention period according to the decisions to be served acts on this second term.

A project to bring to completion?

A first conversation is enough to understand your challenge, identify the right people and tell you frankly what we can bring.