ABB maps growing role for hybrid AC/DC systems

ABB and Boston Consulting Group have set out a case for hybrid AC/DC electrical systems in which alternating current remains the backbone of transmission and much regional distribution while direct current moves deeper into data centres, factories, commercial buildings, and other facilities containing increasingly DC-native loads.

The report, The Strategic Case for Hybrid AC/DC Power: Shaping the Transition to the Next Electrical Architecture, links renewed interest in DC to the growth of AI computing, renewable generation, battery storage, electric vehicles, and industrial automation. Many of those technologies either produce, store, or ultimately consume electricity as DC, even where the surrounding supply architecture remains predominantly AC.

That creates repeated conversion stages. Photovoltaic generation and batteries can pass through inverters before entering an AC distribution system, while servers, electronic loads, variable-speed drives, and other equipment may then convert the electricity back to DC internally. ABB and BCG argue that removing selected conversion stages can improve efficiency and increase the amount of usable power available behind a constrained connection.

The proposal is not to replace established AC networks wholesale. Existing transmission and distribution systems are built around mature AC transformers, switchgear, protection schemes, standards, and operating practice, and the report expects AC to retain that role. The emerging design question is instead where the boundary between AC and DC should sit inside facilities containing large quantities of power electronics.

AI data centres are among the clearest examples because compute density is increasing more quickly than many sites can expand their electrical footprint. ABB and BCG identify 800VDC distribution as an emerging architecture for next-generation AI infrastructure, allowing large blocks of power to be moved through a facility at lower current than would be required at a lower distribution voltage.

Lower current for a given level of power can reduce resistive losses and conductor requirements, although adopting a higher-voltage DC bus introduces its own engineering demands. Protection, isolation, earthing, switching, monitoring, and fault coordination all have to be designed specifically for DC operation rather than borrowed uncritically from conventional AC installations.

Those requirements are becoming a larger part of the 800VDC discussion as suppliers move beyond architecture diagrams and begin developing the equipment needed between the grid connection and the computing rack. Circuit breakers, busway, converters, uninterruptible power systems, distribution equipment, and monitoring platforms must work as a coordinated system if operators are to achieve higher power density without compromising protection or maintainability.

ABB and BCG identify fragmented standards as one of the main obstacles. Competing voltage conventions, interfaces, and protection approaches can leave customers facing proprietary systems whose components are difficult to interchange or expand. Greater harmonisation would give manufacturers a more stable basis for production while allowing engineering companies and operators to develop repeatable design and maintenance practices.

Skills form the second constraint. Engineers experienced primarily in AC systems need appropriate understanding of DC fault behaviour, switching, isolation, and safe working if higher-power installations are to move into routine deployment. The report argues that workforce development needs to progress alongside the technology rather than follow after equipment has already reached sites.

ABB’s interest in DC extends well beyond data centres. The company has used DC-based onboard electrical systems in marine applications for more than a decade and says early installations achieved fuel savings of up to 27% in suitable vessel operating profiles. It launched a solid-state circuit breaker in 2022 and holds more than 700 patents related to DC technologies.

Those examples establish that DC distribution itself is not new. What is changing is the number of applications in which generation, storage, and load are all increasingly electronic and therefore create a stronger case for retaining DC through more of the electrical chain.

Factories can present a similar opportunity where onsite photovoltaics, battery storage, variable-speed drives, robotics, and automated production equipment operate within the same electrical estate. Commercial buildings with substantial electronics or energy storage may also support DC subsystems, although the benefit depends on the actual load mix and the amount of existing AC equipment that would otherwise need replacing.

Retrofit economics are therefore likely to differ sharply from greenfield design. A mature industrial site containing established AC motors, transformers, and switchgear cannot be assumed to benefit simply because some newer equipment is DC-native. New facilities have greater freedom to position conversion stages around the load profile from the outset.

The report places particular emphasis on decisions made during the next two to three years, when suppliers, standards bodies, customers, and engineering organisations are likely to establish many of the conventions surrounding larger DC systems. If interfaces remain fragmented, deployment may proceed through isolated proprietary architectures; if they converge, manufacturers have a clearer route towards repeatable equipment platforms.

The resulting electrical system is unlikely to be purely AC or purely DC. Transmission can remain AC while selected parts of a data centre, factory, microgrid, or charging site operate on DC where conversion losses, power density, or integration with storage justify it. The engineering challenge is to decide where that boundary produces a measurable benefit and then protect it properly.