Three Phases of SMR Control System Innovation (2012–2025)
SMR control system patents and literature span 2012 to 2025, revealing a clear three-phase evolution from foundational hardware experiments to commercial-stage IP formation. Small Modular Reactors are advanced nuclear power units rated at or below 300 MWe, designed for factory manufacture, modular deployment, and enhanced passive safety — and their control systems face demands that have no direct precedent in large-reactor licensing.
The Foundational Phase (2012–2018) established SMR design parameters relevant to control: passive safety features, soluble-boron-free reactivity management, and first digital instrumentation and control (I&C) experiments. Korea Hydro & Nuclear Power’s compatible risk monitoring system (KR, 2012) and General Electric’s self-test protection system architecture (DE, 1984) — still referenced by modern digital protection systems — bracket the early hardware-level work.
The Development Phase (2019–2022) shows accelerating digitalization. North Carolina State University published the Nearly Autonomous Management and Control (NAMAC) system in 2021. Indonesia’s National Nuclear Energy Agency developed FPGA-based reactor protection systems for the RDE reactor in 2021. Ontario Tech University demonstrated real-time SMR simulation in-the-loop for hybrid energy systems in 2022. The Shanghai Key Laboratory published fuzzy control design for SMR once-through steam generators in 2022.
The Maturity and Commercialization Phase (2023–2025) is defined by concentrated patent activity. The dataset contains 6 active or pending patents filed in 2024–2025, indicating that the field is in active commercial-stage IP formation. Korea Hydro & Nuclear Power filed a cluster of multi-module MCR patents in late 2025, covering accident classification, transient-state response, AI-assisted console selection, and scalable facility configurations. CGN Power Co., Ltd. filed a boron-free control rod grouping patent (EP, 2025), and NuScale Power holds an active EP patent for multi-modular power plant electrical grid architecture (2022).
This landscape is derived from a targeted set of patent and literature records retrieved across focused searches. It represents a snapshot of innovation signals within this dataset only and should not be interpreted as a comprehensive view of the full industry. Jurisdiction gaps — particularly the absence of Chinese-jurisdiction patent filings on pure control systems despite Chinese literature activity — may reflect dataset sampling limitations.
Multi-Module MCR Architecture: The Defining Patent Cluster
The most concentrated patent cluster in this dataset addresses a problem unique to SMRs: how a single control room manages multiple reactor modules simultaneously with a reduced operator complement. This multi-module main control room (MCR) architecture challenge has no direct precedent in large reactor licensing, and the IP being filed now will define the regulatory anchor points for years to come.
Korea Hydro & Nuclear Power (KHNP) filed 5 active or pending SMR multi-module MCR patents in the KR jurisdiction in 2024–2025, covering accident classification, AI-assisted console assignment, transient-state isolation, and facility configuration — the most concentrated IP cluster in this dataset.
KHNP’s 2024 filing describes a single MCR with multiple operator consoles, auxiliary large display panels (LDPs), and safety display and control modules, allowing simultaneous monitoring and control of all SMR module operating states. The 2025 patent on facility configuration specifies the quantitative benchmark: N safety display and control modules (one per SMR module) paired with 3×N or fewer operator consoles. This ratio directly addresses the regulatory challenge of minimum-crew multi-module operation.
A companion 2025 patent introduces transient-state isolation logic: when a transient state is detected in one SMR module, the first operator console is isolated to control only the affected unit, preventing cross-module interference. This is a safety-critical design decision with direct implications for how other SMR developers must architect their own MCR systems to avoid infringing KHNP’s pending rights — or to design around them.
“With 5 coordinated pending KR patents covering accident classification, AI-based console assignment, transient isolation, and facility configuration filed simultaneously in late 2025, KHNP may establish foundational rights over the multi-module MCR paradigm in the Korean jurisdiction.”
NuScale Power’s active EP patent (2022) addresses the electrical infrastructure layer of multi-module operation, connecting multiple reactor modules to both a distributed grid and a dedicated service grid through a shared switchyard. This complements the MCR architecture layer by establishing IP at the grid-interface level — a different but adjacent freedom-to-operate consideration for SMR developers in European markets.
Analyse KHNP’s MCR patent cluster and map freedom-to-operate risks for your SMR programme.
Explore SMR Patent Data in PatSnap Eureka →AI-Assisted Management and Digital Twins: Mature Science, Sparse Patents
AI integration into SMR control is academically mature but underrepresented in the patent dataset — a gap that represents a significant IP opportunity for technology developers. The most advanced AI application found in this dataset is KHNP’s 2025 patent for an AI-native MCR architecture, in which the control unit self-learns accident data across SMR units and uses machine learning to classify accidents and determine optimal response console assignment in real time.
North Carolina State University’s NAMAC (Nearly Autonomous Management and Control) system, published in 2021, uses a three-layer architecture — knowledge base, digital twin layer, and operational layer — and was demonstrated on the Experimental Breeder Reactor II during a loss-of-flow accident, with the digital twin acquiring and transmitting operational knowledge to the autonomous controller.
JSC Afrikantov OKBM’s supercomputer digital twin work (2020) applied to the RITM-200 reactor plant combines 1D and 3D coolant circuit modelling to support operational decision-making. Both the NAMAC system and the RITM-200 supercomputer twin represent real-time digital twins as persistent operational tools — not just design aids — that provide anomaly detection, scenario recommendation, and training data generation continuously during plant operation. According to IAEA guidance on digital I&C for nuclear power plants, this class of operational digital twin is an emerging area requiring specific regulatory treatment.
The strategic implication is direct: AI and digital twin integration is well-documented in the literature (2020–2021) but appears underrepresented in the patent dataset, suggesting an IP gap that technology developers could exploit through targeted patent filings on AI-based anomaly detection, digital twin interfaces, and autonomous recommendation engines for nuclear plant operations. As grids increase renewable penetration and OECD NEA member states advance SMR licensing frameworks, the commercial value of this IP gap will only grow.
AI and digital twin integration for SMR control is academically mature — NAMAC (2021) and the RITM-200 supercomputer twin (2020) are both well-documented — but this technology cluster appears underrepresented in the patent dataset. This represents an exploitable IP gap for organisations developing AI-based anomaly detection, digital twin interfaces, and autonomous recommendation engines for nuclear plant operations.
Reactivity Control and Load-Following: The Grid Integration Imperative
Load-following capability is identified as a primary operational requirement for SMRs competing with renewables, and the patent and literature activity in this cluster reveals both the technical approaches available and the IP gaps that remain. The KEPCO International Nuclear Graduate School review (2022) identifies five comprehensive load-following requirements for SMRs: planned operation, automatic generation control, governor-free operation, cooperative reactor-turbine control, and multi-module unit control.
Clear Inc.’s active EP patent (2021) for a load-following liquid-metal-cooled SMR uses a neutron reflector coupled to a thermally expanding liquid or gas mechanism that naturally adjusts reactivity with heat output, enabling passive load-following without active control rod adjustment — a novel architecture for liquid-metal SMRs.
CGN Power Co., Ltd.’s EP patent (2025) introduces functionally grouped control rods — power compensation, reactivity adjustment, temperature adjustment, and shutdown rods — configured to eliminate soluble boron entirely, with insertion depth as the sole reactivity management mechanism. This boron-free architecture, also examined in the KEPCO KINGS reactivity balance study (2018), transfers all reactivity management responsibility to rod control algorithms and interlock systems, imposing tighter requirements on rod position measurement, worth calculation software, and fail-safe interlock architectures.
The University of South China’s dual fluid reactor control scheme (2021) — designed for a molten salt plus liquid lead reactor — incorporates uncertainty quantification from model and material property variations directly into controller design, achieving robust performance across parameter ranges. The Shanghai Key Laboratory’s variable universe fuzzy controller (2022) adapts PI parameters dynamically in response to steam pressure and superheat error signals for the secondary circuit of SMR once-through steam generators. Both approaches address the reality that SMR control systems must perform across wider operating envelopes than traditional large-reactor designs, a point underscored by IEA analysis of nuclear flexibility requirements for high-renewable grids.
The SMR–hydrogen cogeneration control challenge adds a further dimension. EGB Engineering’s analysis (2022) identifies the need for power-splitting control logic that dynamically redirects thermal output between electrical generation and hydrogen production based on real-time grid pricing and demand signals — a control architecture with no precedent in traditional nuclear plant design. This bidirectional setpoint management introduces new I&C design requirements not addressed by any patent in this dataset, representing another white-space IP opportunity.
Map white-space IP opportunities in SMR load-following and hydrogen cogeneration control with PatSnap Eureka.
Analyse Patent White Space in PatSnap Eureka →FPGA Protection Systems and Cybersecurity by Design
FPGA-based reactor protection systems represent the hardware transition from legacy analog platforms to programmable digital I&C, and the cybersecurity implications of this transition are now a regulatory imperative. Indonesia’s Pamulang University designed and modelled an FPGA-based reactor protection system using CompactRIO for the Reaktor Daya Eksperimental (RDE) in 2021, selected for high reliability, reprogrammability, and fast loop execution rates, validated against postulated initiating events from the RDE design basis.
GE-Hitachi’s active EP patent (2020) for a nuclear reactor scram control system describes a solenoid pilot valve system with integrated indicator lights providing immediate visual confirmation of scram system energisation states, reducing operator radiation exposure and inadvertent scram risk. General Electric’s 1984 DE patent — now inactive — established the foundational self-test architecture in which a central microprocessor serially addresses channel cards and verifies functional continuity via short, system-transparent test pulses. This conceptual basis is still referenced by modern digital protection system self-diagnostic functions, illustrating how foundational I&C IP persists across technology generations.
The Canadian Nuclear Safety Commission’s 2020 regulatory paper on SMR security identifies that SMR digital I&C platforms expand the cyber attack surface compared to analog predecessors, and calls for a graded, risk-informed Security by Design (SeBD) approach to be embedded into SMR licensing from the first design iteration.
The Canadian Nuclear Safety Commission’s SeBD framework (2020) and the expansion of digital I&C platforms together create a mandatory design requirement that will become a licensing condition in multiple jurisdictions. According to NRC guidance on cybersecurity for digital I&C systems, the attack surface introduced by networked operator consoles and FPGA-based protection systems requires defence-in-depth architectures from the earliest design stage. Organisations building SMR control systems must treat cybersecurity architecture as an I&C design constraint from the first design iteration, not a post-design overlay.
Security by Design (SeBD) is a regulatory approach — identified by the Canadian Nuclear Safety Commission (2020) — that integrates both physical and cyber security requirements into SMR licensing from the earliest design phase, rather than applying them as a post-design overlay. The Canadian Nuclear Safety Commission identifies a graded, risk-informed approach as the appropriate framework for SMR digital I&C platforms.
Strategic Implications for IP and R&D Teams
Five strategic implications emerge directly from the patent and literature evidence in this dataset, each with actionable consequences for IP strategists and R&D leaders working on SMR programmes.
KHNP’s MCR cluster as a potential freedom-to-operate chokepoint
With 5 coordinated pending KR patents covering accident classification, AI-based console assignment, transient isolation, and facility configuration filed simultaneously in late 2025, KHNP may establish foundational rights over the multi-module MCR paradigm in the Korean jurisdiction. Competitors entering this space — particularly those seeking to license Korean SMR technology — should assess freedom-to-operate against this cluster before finalising their MCR architecture decisions.
Boron-free architectures demand I&C investment
Eliminating soluble boron transfers all reactivity management responsibility to rod control algorithms and interlock systems. R&D teams designing SMR I&C platforms must invest in high-resolution rod position measurement, enhanced worth calculation software, and fail-safe interlock architectures that were previously shared with chemical reactivity control. This is not an incremental upgrade — it is a fundamental redesign of the reactivity management layer.
Load-following control software is the critical IP white space
As grids increase renewable penetration, SMR operators will face mandatory load-following requirements. Clear Inc.’s passive thermal-expansion reflector mechanism (EP, 2021) and the KEPCO KINGS load-following requirements review (2022) together define the technical and regulatory envelope. IP strategists should focus on the control software layer — automatic generation control algorithms, governor-free operation logic, and cooperative reactor-turbine coordination — where patent activity remains sparse in this dataset. The WIPO Green technology patent database confirms that nuclear load-following control software remains a relatively uncrowded IP space compared with adjacent renewable energy control technologies.
Remote and marine operation create additional I&C design constraints
Multiple literature results identify SMR control systems designed for autonomous or semi-autonomous operation in remote, isolated, or Arctic environments. The Russian RITM designs and the Holos-Quad micro-reactor concept — which fits into a standard ISO shipping container — both require control architectures capable of extended operation without on-site specialists. Marine propulsion applications (Cockcroft Institute, 2019; Military Institute of Science and Technology, 2018) add the elimination of soluble boron chemical control during power operation and small crew size as additional driving constraints, making control rod worth the sole reactivity management mechanism.
Cybersecurity is a licensing condition, not a design option
The Canadian Nuclear Safety Commission’s SeBD framework (2020) and the expansion of digital I&C platforms together create a mandatory design requirement that will become a licensing condition in multiple jurisdictions. The dataset shows that FPGA-based protection systems, networked operator consoles, and AI-assisted MCR architectures all expand the cyber attack surface. Organisations building SMR control systems must treat cybersecurity architecture as an I&C design constraint from the first design iteration.
Within the SMR control system patent dataset analysed, jurisdiction concentration is: KR (5 patents), EP (4 patents), US (2 patents), DE (1 patent, legacy). Research institutions — including North Carolina State University, Ontario Tech University, Shanghai Jiao Tong University, University of South China, and Indonesian institutions — contribute the majority of literature-based innovation signals on autonomous control, digital twins, and FPGA-based protection systems.
The overall picture is of a field transitioning rapidly from academic proof-of-concept to commercial IP formation. The 6 active or pending patents filed in 2024–2025 represent the leading edge of that transition. For IP and R&D teams, the window for establishing foundational positions in AI-native MCR architecture, load-following control software, and digital twin operational interfaces is open — but narrowing. PatSnap’s innovation intelligence platform, used by 18,000+ customers across 120+ countries, provides the patent analytics infrastructure to identify these gaps and act on them before the field consolidates.