DERMS market forecast at $7.04 billion by 2035 as data centre loads pile pressure on grids
Grid software spending accelerates as utilities struggle with intermittent renewables and surging power demand.
The global distributed energy resource management system market is projected to reach $7.04 billion by 2035, according to a report published Wednesday (2026-07-29) by Market Research Future, as grid operators grapple with rising renewable capacity and data centre power loads. The market, valued at $1.78 billion in 2026, is expected to expand at a compound annual growth rate of 16.5% through the forecast period.5
The physical limits of existing grids have become the dominant constraint for both generators and large consumers. Without DERMS software to orchestrate distributed solar, battery storage and demand response, utilities risk daily violations of frequency and voltage tolerances.
Asia-Pacific is projected to register the fastest growth, with a CAGR of 19.2%, according to the same report. India’s 500 GW non-fossil capacity target, China’s smart grid investment programmes and renewable expansion across Southeast Asia are pushing distribution networks beyond original design capacity.5
The timing aligns with a separate forecast from Persistence Market Research, published in May (2026-05-26), which pegged the AI in energy distribution market at $42.7 billion by 2033. North America dominates that segment with roughly 30% share, but East Asia — holding around 22% — is the fastest-growing region, driven by industrial digitisation and smart city initiatives in China, Japan and South Korea.1
The United States alone accounts for 45% of global data centre electricity consumption, with demand expected to rise by nearly 240 terawatt-hours — a 130% increase — according to the same Persistence report. China contributes another 25% of global data centre power use, with projected growth of 175 terawatt-hours, or 170%. That concentrated load growth creates acute grid stress that regulators cannot solve with new transmission alone.1
Utilities are responding faster than many expected. The Persistence report noted that 41% of utilities have already fully integrated AI and analytics systems ahead of schedule, with companies like ABB Ltd and GE Vernova advancing AI-driven energy platforms.1
The broader context is a capital cycle that could dwarf anything seen in the past decade. Analysts at Bloomberg Intelligence, cited in a separate forecast from June (2026-06-04), project that extreme weather will drive more than $20 trillion in global spending over the next decade, with the capital reallocation shifting from reactive disaster recovery to preventive infrastructure hardening.3
That framing recasts DERMS software not as a niche automation tool but as a prerequisite for any utility hoping to keep the lights on during summer heatwaves or winter storms. The smart energy market is projected to reach $417.7 billion by 2033, up from $156.1 billion in 2026, according to a PRNewswire report from March (2026-03-09).2
The smart electricity meter segment alone is forecast to grow from $15.02 billion in 2026 to $23.46 billion by 2031, a CAGR of 9.31%, according to Mordor Intelligence data published in June (2026-06-10).4 That metering build-out provides the granular data DERMS platforms need to function.
Still, the growth forecasts in these reports rely on an assumption that regulatory frameworks will keep pace with technology, which has not been the case in many jurisdictions. Grid interconnection queues in the US and Europe remain backlogged for years, and permitting reform remains stalled. The DERMS market will only hit its high-end projections if utilities can actually deploy the software on networks still running copper and analogue switches.
The number to watch is not the 2035 market size but how many utilities move from pilot programmes to full system-wide deployment in the next 12 to 18 months. If the 41% integration rate accelerates, installed capacity will force regulators to act. If it stalls, the software will sit on servers while transformers overload.