Multi-physics simulations across the whole device lifecycle — feasibility, design, engineering, operations — proven from small university devices up to ITER scale and complexity.
From single sensors to the full Modular Diagnostics Platform — validated measurements feeding plasma control and machine safety on next-generation MCF devices.
Plasma control and operations optimization — classical controllers strengthened with reinforcement learning and other AI/ML techniques, validated on operating devices.
Checked against experimental data from multiple tokamaks, benchmarked against established codes, and reviewed by external subject-matter experts.
Conventional and RL controllers trained in our framework, tested on DIII-D and other devices — another confirmation of the framework's accuracy.
Demo and public API free on FusionTwin.io; on-premise and custom integrations on request.
DINA: R.R. Khayrutdinov & V.E. Lukash, Studies of Plasma Equilibrium and Transport in a Tokamak Fusion Device with the Inverse-Variable Technique, J. Comput. Physics 109, 193 (1993) · NSFsim at DIII-D: R. Clark et al., Validation of NSFsim as a Grad-Shafranov Equilibrium Solver at DIII-D, Fusion Eng. Des. (2025), arXiv:2412.03786
Blog series: NSFsim Perspective on Disruptions in Tokamaks — Part I: Physics Basis · Part II: Simulations for DTT — blog.nextfusion.org (2025)
Paper: S. Guizzo et al., Electromagnetic System Conceptual Design for a Negative Triangularity Tokamak, arXiv:2501.14682 (2025) · Blog: NTT: Conceptual Design · NTT: Preliminary Design
Engage us as early as possible to get proven technologies and processes deployed, the groundwork done, and the road ahead paved — we help build your team, transfer the work, and stay engaged long-term.
We work alongside your existing team, verifying and validating your tools and results, and supporting the most complex and sophisticated research — extra depth exactly where your team needs it.
Deploy the NSFsim framework in-house — your infrastructure, your data, your workflows — under flexible licensing, from free research collaborations to full commercial terms with IP transfer and support.
A diagnostic is only as reliable as its supply chain. We actively manage ours — component quality, repeatability, long-term availability, easy maintenance and support.
Measurements alone are not enough — we reconstruct missing plasma parameters in real time using PINNs and fast solvers, delivering a complete, trustworthy plasma state to control and safety.
ECE, profile reflectometer, CTS, and radial interferometer-polarimeter.
Non-perturbative, no in-vessel calibration targets, reliable in high magnetic field and neutron environments — the core of future FPPs' I&C.
Epitaxial graphene-on-SiC Hall sensors with an intrinsically low neutron-damage cross-section and thermally activated self-healing — built for FPP conditions: D-T neutron fluences of 10²⁰–10²² cm⁻² and temperatures up to 770 K.
Further diagnostic types are on the roadmap — and we constantly grow our network of partners and suppliers to cover the demand of next-generation MCF devices and future FPPs.
Let's discuss!
Our vision for MDP: all measurements required by plasma control and device safety — provided in real-time by a minimalistic, 100% FPP-relevant set of sensors.
Modular port-plug assembly consolidating all front-end hardware — radiation-hardened, with neutron & gamma shielding, thermal management, and vacuum boundaries.
Versatile Integrated Data Acquisition system: acquisition & control boards, industrial chassis, networks, and software — one platform for native and third-party sensors.
Validated, high-integrity real-time data delivered straight to plasma control and device safety systems — hybrid model-based and ML reconstruction with uncertainty quantification throughout.
Standardized, off-the-shelf, fast-replaceable components backed by a purpose-built, actively managed supply chain — controlled costs, guaranteed availability, and long-term maintenance.
Factory acceptance testing, on-site commissioning, and calibration, followed by long-term operational support — covering the full device lifetime from first plasma to decommissioning.
Self-healing: S. El-Ahmar et al., Appl. Surf. Sci. 685, 161953 (2025) · Thermal stability: T. Ciuk et al., IEEE EDL 45, 1957 (2024)
We develop controllers combining conventional approaches with machine learning and reinforcement learning to achieve superior robustness and keep control transparent.
Full discharge scenarios — initiation, ramp-up, flat-top, and ramp-down — optimized across energy, stability, and operational goals, with actuator limits and stability boundaries respected by design.
We aim to provide a plasma control system for MCF devices and future FPPs that combines all of our expertise — integrated modeling, diagnostics, scenario development, and control.
RL control at DIII-D: G.F. Subbotin et al., Demonstration of reconstruction-free static magnetic control of DIII-D plasma with deep reinforcement learning, Nucl. Fusion (2026) · Sensor-robust shape control: D. Sorokin et al., Dynamic Plasma Shape Control with Arbitrary Sensor Subsets, arXiv:2605.15935 (2026)
Blog: Controlling Plasma Temperature and Safety Factor with Gyrotrons Using Reinforcement Learning — blog.nextfusion.org (2026)
100% of our work is dedicated to fusion energy — rooted in deep plasma physics expertise and active engagement with the fusion community.
Integrated modeling, diagnostics, plasma control, and AI/ML under one roof — covering the full I&C stack from first sensor to final actuator.
We actively manage our supplier and partner network — ensuring component quality, repeatability, long-term availability, and easy maintenance.
Decades of expertise in plasma modeling and experimental validation. We develop NSFsim and contribute to community codes such as TRAVIS.
Models and results rigorously verified and validated against experimental data from multiple tokamaks — together with leading physicists and engineers.
The earlier we engage, the better the outcome — from the very first tokamak concept. Don't wait until construction begins.
Entrepreneur and investor, dedicated to fusion.
Senior plasma physicist; Kurchatov, ITER.
Integrated modeling expert; JET, EUROfusion, TE.
Physics, software, product background.
Plasma physicist; tokamaks, stellarators.
Microwave, laser sensors; JET, WEST, ITER.