Idle Substation Sites Become ESS Battery Sites — How KEPCO’s K-RE100 and the SMR Site in Gijang County Are Reshaping the Grid

① What’s the Story

Two decisions made in Korea’s power infrastructure in the summer of 2026 involve entirely different power sources, yet point in the same direction: instead of looking for new land, reuse existing power sites where permitting and grid connection are already complete. Korea Electric Power Corporation (KEPCO) has moved into full swing with its K-RE100 program to build a total of 95MW of solar on idle land at 500 substations nationwide by 2030, and Gijang County in Busan has been selected as the site for Korea’s first small modular reactor (SMR, Small Modular Reactor), pursued by Korea Hydro & Nuclear Power (KHNP). For the secondary battery industry, this trend can be read as a signal that the way “land” is sourced for ESS (Energy Storage System) battery racks is changing.

Starting with the industry situation: power demand keeps rising as AI data centers spread and EV charging infrastructure grows, but new generation and transmission facilities take years to reach construction because of permitting and local acceptance issues. The tension is that securing sites has itself become the bottleneck. Renewables, ESS, and nuclear projects alike often stall in front of the question, “where do we build it?”

This article answers three questions. First, in what way, and to what extent, can reusing existing substation and power plant sites relieve this bottleneck? Second, what do the published figures of the two projects tell ESS battery businesses? Third, from a patent perspective, where should planning and IP personnel be looking?

② Basic Principle

Using idle substation land can be compared to building a house on a lot where roads, water, and sewage lines are already in place. A substation already hosts 154kV-to-345kV-class transmission lines, switchgear, and transformers, so placing solar panels or ESS containers in the empty space within it allows grid connection without building new transmission lines. It is a structure that cuts both the cost and the time of grid interconnection at once.

KEPCO’s K-RE100 (Korean-style RE100, a program aiming at 100% renewable sourcing in the public sector) plan follows this principle directly. According to a July 28, 2026 report, KEPCO signed an agreement under which it leases idle substation land while KEPCO Solar installs and operates the equipment. In the first phase, 4.8MW will be built at five substations: 1.8MW at two sites including the Banwol substation in Ansan, Gyeonggi Province, by December 2026, and roughly 3MW more at three sites including the Seongeo substation in the first half of 2027, with expansion through competitive bidding from 2027 onward.

The SMR site selection follows the same principle. An SMR is a reactor with an electrical output of 300MW or less, fabricated in modules at a factory and assembled on site, and it needs less land than a large nuclear plant. In the results announced by the New Nuclear Power Plant Site Selection Evaluation Committee on June 17, 2026, Gijang County scored 87.11 points, ahead of Gyeongju (84.56 points). The reported reasons were that the Gori Nuclear Power Headquarters contains about 220,000㎡ of idle, already-prepared land allowing immediate construction, and that the existing transmission lines of the permanently shut-down Gori Unit 1 can be reused, removing the need for new transmission towers or land purchases.

• Common principle: reuse existing sites and grid connection points instead of new land and new transmission lines.
• Difference: the substation model “fills empty space,” while the SMR model “uses reserve land inside an existing generation complex.”

③ Comparison

The table below sets out the two projects along six axes — power source, site, scale, timeline, grid connection, and battery-industry relevance — together with figures confirmed in published reports.

Category	KEPCO idle-substation solar	KHNP SMR site in Gijang County
Power source	Solar (renewable energy)	Innovative small modular reactor (i-SMR)
Site characteristics	Idle vacant land inside existing substations	Existing idle land inside Gori Nuclear Power Headquarters (about 220,000㎡)
Project scale (MW)	500 sites by 2030, 95MW total (phase 1: 5 sites, 4.8MW)	One unit, 0.7GW (700MW)
Timeline	1.8MW by Dec 2026 → approx. 3MW added in H1 2027 → competitive bidding thereafter	Standard design approval 2028 → construction start 2030 → commercial operation targeted 2035
Grid connection advantage	Uses existing substation equipment; no new transmission lines needed	Existing Gori Unit 1 transmission lines can be reused
Battery industry relevance	Low-cost site candidate when small ESS units are installed alongside	Interpreted as relatively smaller ESS buffering demand than renewables, given its baseload nature

What this table says: The core advantage of both projects is “reusing existing grid connection points,” and in scale a single SMR unit (700MW) is about 7.4 times all 500 substation sites combined (95MW). From an ESS perspective, the key point is that the substation model is not a large power source but a dense network of small, distributed sites.

Here we run one simple calculation on the published figures. The five phase-1 sites average about 0.96MW, but dividing the remaining 90.2MW across the remaining 495 sites gives an average of only about 0.18MW (about 182kW) per site. In other words, once past the initial pilot sites, per-site installations shrink to the hundreds-of-kW range, so any ESS installed alongside is likely to be a standardized small modular product rather than a large container park. The figures suggest that, for battery businesses, the meaning of this program lies less in “large orders” and more in “small standard products that can be installed repeatedly.”

④ Patent Perspective

Claim gist and scope of rights: The business model itself — leasing idle substation land and installing solar and ESS equipment — is governed less by patents than by site lease agreements and the permitting structure under the Electric Utility Act. Within the scope of this article’s research, no registered patent directly claiming the business model of “reusing idle substation land” could be confirmed from published patent information. This article therefore makes no infringement-risk judgment based on any specific patent number.

Filing trends and tracking points: The layer where rights competition will actually play out is not the land but the equipment placed on it. Fitting batteries into narrow idle substation land makes container-level thermal management, fire-spread prevention, and clearance and protection-coordination design with existing substation equipment critical. Because battery and power-equipment makers file in this area individually, tracking the following items with search queries is the realistic approach.

• Tracking point 1: filings on fire-protection and clearance design for ESS installed inside substation premises (by applicant, via KIPRIS and Google Patents)
• Tracking point 2: claims on grid-interconnection control (peak-discharge scheduling) for combined solar-ESS installations
• Tracking point 3: claims on rack and cooling structures for small modular ESS in the hundreds-of-kW class

Design-around scenario (counterargument): If a business supplying small modular ESS runs into a particular cooling-structure patent, switching between air and liquid cooling or changing the module arrangement could omit one claim element and possibly avoid literal infringement. There is a counterargument, however. The smaller the site, the less design freedom there is, and designs tend to converge on similar layouts and cooling methods. In that case the design-around options narrow and the room for a doctrine-of-equivalents assessment grows, so the more site-constrained a project is, the earlier the FTO (freedom-to-operate) review needs to move into the design stage.

The bidding structure also shifts where IP risk sits. Once operators are chosen through competitive bidding from 2027 onward, liability for patent disputes over the equipment is likely to pass to equipment suppliers, usually through patent warranty and indemnity clauses in supply contracts. ESS suppliers therefore need to sort out whether their products conflict with others’ scope of rights before bidding, so they can avoid unfavorable indemnity terms in contract negotiations.

⑤ Limitations and Outlook

Viewed optimistically, reusing existing infrastructure looks like a way to speed up power infrastructure expansion. But the counter-scenario is also clear. Idle substation land is not large to begin with, so all 500 sites add up to only about 95MW, and as shown above, the average per-site scale shrinks sharply after phase 1. The argument that this approach alone cannot absorb the growth in power demand driven by AI data centers remains valid. The SMR, too, must pass through multiple stages such as standard design approval and environmental impact assessment before its 2035 commercial-operation target, so the possibility of schedule changes should be kept in mind.

Placing the two projects side by side reveals one shared lesson. The reported reason Gijang beat Gyeongju and the reason idle substation land was chosen both came down to “grid connection points already secured.” This suggests that in future competition for ESS sites, the scarce asset may be connectable grid capacity rather than land area. If so, ESS suppliers’ competitiveness may shift toward high-density, high-safety designs that fit more energy safely into the same connection point, and the center of gravity of patent portfolios may likewise move from cell chemistry toward installation and safety design.

As a strategic alternative, one direction worth considering is piloting a “distributed mini-ESS” model, installing small ESS units alongside substation solar to ease local grid strain by discharging at peak hours. Because per-site scale is small and installations are highly repeatable, businesses that secure standard products and installation procedures first may have an advantage over those chasing single large projects.

Practical action: Planning and IP personnel should periodically check KEPCO’s K-RE100 competitive-bidding announcements expected from 2027 onward, and, alongside a specification sheet of installation area and thermal-design margin for hundreds-of-kW small ESS modules, run an FTO screening in advance of valid cooling- and fire-protection-related patents corresponding to those specifications.

Items requiring verification

• The name of the second phase-1 site (1.8MW across two sites) other than the Banwol substation
• Whether combined ESS installation is actually within the scope of KEPCO’s K-RE100 substation program
• The publication number, applicant, and claim scope of the isostatic-press equipment patent discussed in connection with ESS cell manufacturing precision
• Whether any registered patent in Korea or elsewhere directly claims the business model of reusing idle substation land

Sources

1. Newspim, “KEPCO to build a total of 95MW of solar at 500 substations; ‘K-RE100’ in full swing” (2026-07-28)
2. Aju Business Daily, “Gijang County confirmed as final innovative SMR site; ‘immediate construction’ potential beat Gyeongju” (2026-06-18)
3. Kukmin Ilbo, “A future reactor succeeding Gori Unit 1: Gijang embraces the SMR” (2026-06-18)
4. M Economy News, “Yeongdeok and Gijang selected as new nuclear sites; two large reactors and one SMR planned” (2026-06-18)