The 15th five-year plan for the new energy system, published in June 2026 by the NDRC, aims for more than 300 GW of newly distributed renewables by 2030, roughly 60 GW per year. This target covers the full range of distributed deployment scenarios: industry and commerce, transport, buildings, and agriculture. These sites will adjust their consumption in real time to available production. At this scale, the deployment is unprecedented.

The Essentials

  • China targets 3,500 GW of total renewable capacity for 2030, including 300 GW of “new distributed energy” integrated directly into factories, mines, farms, and buildings over 2026-2030 (NDRC, 15th five-year plan, June 2026).
  • The architecture relies on the flexibility of industrial demand rather than centralized battery storage: sites adapt to production, not the other way around.
  • The NDRC coordinates this deployment through dynamic pricing mechanisms and various measures targeting a growing number of major industrial consumers.
  • The bet is new at this scale: no continental grid has yet proven that industrial flexibility can replace massive storage as a balancing mechanism.

60 GW per Year, and a Hypothesis Worth Testing

To grasp the ambition, consider this benchmark: 60 GW per year of distributed renewables is roughly twice Belgium’s total installed electrical capacity, which reached 27.5 to 28.25 GW in 2023. China commits to deploying this annually, for five years, across small-scale installations connected to the local distribution network. These sites can inject generated energy into the grid or consume it locally; they are not exclusively off the centralized network.

The traditional energy transition model rests on two pillars: gigantic solar or wind plants connected to a high-voltage transmission network, and batteries capable of storing the excess produced when the sun shines or the wind blows. This model is proven. It is also costly, demands critical metals, and depends on network infrastructure that many countries cannot afford to modernize rapidly.

The 15th Chinese plan pushes a different logic: bring production closer to the consumption site, then ensure that site adjusts. A factory receiving solar panels on its roof can decide, when production peaks at noon, to intensify certain energy-intensive production lines, and slow them when clouds arrive. A mine can accelerate its water-pumping during hours of surplus. A greenhouse can trigger its irrigation or air-conditioning systems at the most opportune moment. In this scheme, the battery is the factory itself.

The NDRC Builds the Regulatory Framework That Makes Flexibility Possible

The coexistence of regulatory obligations and tariff incentives creates a risk of contradictory signals for industrialists: a participation obligation can reduce the perceived value of flexibility by turning it into an administrative constraint rather than an economic opportunity. This shift alters operator behavior, which then tends to satisfy the minimum requirement rather than optimize their response, weakening precisely the granularity of adjustment that the plan seeks to mobilize.

Such a system works only if prices send the right signals. China is building this framework in parallel with physical deployment.

The plan provides for the expansion of dynamic pricing mechanisms to a significant share of major industries. Concretely: electricity rates vary according to hour, day, and grid state. An industrialist who accepts adjusting their consumption in response to these signals can buy electricity more cheaply during off-peak hours and sell their flexibility to the grid for remuneration. It is an auxiliary services market, modeled on approaches already tested in Europe and California, but never at this scale.

The NDRC intends to strengthen demand response obligations for a growing number of large industrial consumers. Voluntarism is insufficient: sites beyond a certain consumption threshold will have to equip themselves with the necessary control systems and participate in flexibility mechanisms. The goal is to create a critical mass of controllable sites before the grid is overwhelmed by renewable production variations.

This coupling of physical deployment and incentive framework is the heart of the gamble. Panels and turbines alone are worthless if industrialists have no economic reason to change their behavior. Dynamic prices alone are insufficient if sites are not equipped to respond. The plan attacks both fronts simultaneously.

Heavy Industry as Priority Target

Not all industrial consumers are equally suited for this exercise. China concentrates its attention on three types of sites that meet favorable conditions.

Mining and extractive industries first. These sites operate with massive electrical consumption—ventilation, pumping, grinding—and some of these operations are inherently flexible. Ore grinding can be shifted an hour without affecting daily output. These sites are often far from urban centers, making them difficult to supply via the centralized grid and natural candidates for self-generation.

Manufacturing industry next, particularly sectors operating on three shifts that can modulate the intensity of certain lines without stopping production. Aluminum, cement, basic chemicals: sectors where energy represents an enormous share of production cost, creating strong economic motivation to adapt to dynamic prices.

Industrial agriculture finally, notably heated and air-conditioned greenhouses, intensive livestock operations, large-scale irrigation systems. China’s agricultural sector is undergoing full transformation toward more intensive and automated models, and these models consume significant electricity at predictable times.

The geographic concentration of these industries in certain provinces allows the plan to target pilot zones before nationwide generalization. It is a graduated approach, characteristic of Chinese planning methodology: experiment at the provincial scale, measure, correct, then deploy nationally.

Lessons from California and Europe

Industrial flexibility as a grid-balancing mechanism is not a Chinese idea. It has been tested for several years in California and certain European countries. The results are instructive.

The California Independent System Operator (CAISO) documented between 2023 and 2025 the scaling up of industrial demand response programs. These programs contributed to grid balance during summer peak demand, but CAISO’s primary sources do not confirm they avoided blackouts. Real performance during the most intense peaks fell short of contractual capacities. Their contribution remains limited by the number of sites actually controllable and by some industrialists’ resistance to modifying production processes. Flexibility has a real organizational cost for companies, and this cost sometimes exceeds the economic gains from dynamic pricing.

In Europe, the most advanced experiments are in Scandinavia and Germany. Electricity-intensive industries—aluminum, steel, paper—actively participate in flexibility markets in these countries. But their contribution remains marginal at the network scale, because these industries themselves are in decline or undergoing difficult competitive transformation.

The Chinese difference lies in scale and development phase. China deploys this architecture precisely when its industry is in full expansion, not decline. New factories can be designed from the outset to integrate flexibility: sensors, automated control systems, connections to electricity markets, without retrofitting aging installations. It is a first-mover advantage that Europe no longer possesses.

The Unproven Hypothesis at the Plan’s Center

The rigidity of industrial processes constitutes the main uncertainty factor, as certain production lines do not tolerate even brief interruption without significant restart costs. This tolerance threshold varies widely from sector to sector and directly conditions the share of demand actually controllable. The more constraining this threshold proves, the further the volume of flexibility effectively mobilizable diverges from the theoretical potential on which the plan’s architecture rests.

The plan rests on a hypothesis not yet verified at this scale: distributed industrial flexibility can serve as the primary balancing mechanism for a continental grid, in place of centralized storage.

The question deserves precise formulation. Battery storage has a simple virtue: it is passive. You charge when energy is available, you discharge when it is lacking, without asking an industrialist to modify their processes. Industrial flexibility, by contrast, supposes that site operators accept allowing external algorithms to influence their production decisions. It is a managerial change as much as a technical one.

The plan’s first years will therefore be a full-scale test. If Chinese industrialists actually respond to price signals, if automated control systems deploy at the planned pace, and if renewable production variations remain within manageable ranges, the architecture could hold. If industrial demand proves more rigid than expected, because processes do not tolerate variations, because economic incentives are insufficient, or because control systems struggle to deploy, China will probably have to combine this approach with massive storage.

The IEA, in its work on grid modernization published in 2025, emphasizes that the most resilient systems combine multiple balancing mechanisms rather than a single one. The Chinese plan also provides for centralized storage, but bets that distributed flexibility can significantly reduce its need and thus its cost.

This shift in priority raises another question: that of data. A distributed flexibility system at this scale generates enormous volumes of information on industrial behaviors, production processes, real-time consumption. The centralization of this data by grid operators, and ultimately by the State, creates a level of visibility over China’s industrial economy without equivalent elsewhere. It is a control benefit and an informational concentration of power that the plan does not mention explicitly.

A Laboratory Whose Data Will Belong to Everyone

The 15th Chinese energy plan is first and foremost a national policy, with its decarbonization, supply security, and industrial competitiveness objectives. But its results will be of interest far beyond China’s borders.

Developing countries building their electrical grids today are watching this model closely. Many of them possess abundant solar or wind resources, nascent industry that can be designed from the outset for flexibility, but lack means to invest in giant batteries or high-voltage lines. If China demonstrates that distributed industrial flexibility works at large scale, it offers an alternative model, and probably less costly, that these countries can adapt.

For advanced economies, the stakes are different. They already have aging networks, industries difficult to retrofit, and liberalized electricity markets where coordination is more complex. But the signals emerging from Chinese provinces between 2026 and 2030—actual industrial participation rates, implementation costs, flexibility levels actually achieved—will provide data that neither CAISO nor European operators have been able to generate at this scale.

The Chinese institutional context, the NDRC’s capacity to impose obligations on industrialists, and the vertical integration of certain actors are not found in other systems, which limits the transposability of results. By 2030, this plan will allow measuring what portion of observed flexibility stems from technical architecture and what portion stems from the State’s capacity to impose behaviors that markets alone would not produce.

The answer will say as much about energy transition feasibility as about the political conditions that make it possible.

Sources

  1. NDRC 15th Five-Year Plan for New Energy System, analyzed by Carbon Brief: https://china-insights.org/2026/07/chinas-15th-five-year-plan-for-a-new-energy-system/
  2. IEA, Grid Modernisation and the Smart Grid, report 2025, iea.org
  3. CAISO, annual demand response reports 2023-2025, caiso.com