Skip to content
See the World Through ScienceA project of ALLATRA
Source: Peer-reviewedCommunications Engineering1 source

Study Finds Data Centers Can Help the Grid, but Only if They Can Prove It

AI & Technology

Republish this story

Our work is licensed under Creative Commons BY-NC 4.0. You may republish this piece for free — with credit to ALLATRA Media and a link to the original, unedited beyond length trims, and not for commercial use.

Read the full license

An electrical substation with transformers, steel gantries and insulators, with a transmission pylon and open ground behind it.
A substation in the Nievre, central France, photographed Oct. 1, 2015 (illustrative). The study sets a separate electricity price at each point of a distribution network like this one, hour by hour."Electrical substation, Cosne-Cours-sur-Loire, 58-Fr (2)" by Cjp24, via Wikimedia Commons, CC BY-SA 4.0

Researchers at four universities in China have published a model that sets electricity prices on a local distribution grid hour by hour and shifts computing work between data centers in response. In their simulations, signing a demand-response contract, which pays large power users to cut consumption on request, is worth it only when the promised cut can be delivered and verified.

The authors describe data centers as both a fast-growing, high-density load on distribution networks and an unusually movable one, since a job can often be run in another building or at another hour. Their framework is an attempt to turn that flexibility into reductions a grid operator can count.

The model works at two levels: the network operator sets the prices, and the data centers reschedule their jobs across sites and hours in response. The team built it around the demand-response rules in force in Guangdong, China, which set thresholds a response must meet and penalize responses that do not count as valid.

The paper reports three results from the simulations: lower data-center costs, more renewable generation taken up, and more of the promised flexibility counting as a valid response. The test cases were a modified IEEE 32-bus system, a standard research test grid, and a 60-bus system the authors describe as practical. Neither is a live utility network.

The authors conclude that "under threshold-based settlement, participation in the demand response is not necessarily beneficial unless flexibility can be delivered in a valid and verifiable form."

The study was published Nov. 21, 2026, in Communications Engineering. Zhixiang Xing of The Chinese University of Hong Kong (Shenzhen) is its first author, and Hongbin Sun of Taiyuan University of Technology is listed last. The version online is the peer-reviewed accepted article, which the publisher says will be replaced by the final edited one.

The paper lists funding from the National Natural Science Foundation of China and from Shenzhen and Hong Kong programs, and declares no competing interests.

Sources

By Olga SchmidtEditor-in-Chief, Writer

Spot an error?

Spot an error?

Report an error

Spotted a mistake on this page? Tell us what's wrong and our editors will take a look.

What kind of problem?

Only if you'd like us to be able to follow up. We won't use it for anything else.

We correct mistakes openly. Select any text to flag it. Fixes are logged under our Corrections Policy.

Report an error

Reporting on

Study Finds Data Centers Can Help the Grid, but Only if They Can Prove It

What kind of problem?

Only if you'd like us to be able to follow up. We won't use it for anything else.

We read every report. Corrections are logged publicly.