Skip to content
Preprint

Reinforcement learning for vertical position control on the EXL-50U spherical tokamak

Aug 2026 · 1 citation · 32 references
Physics

Abstract

Vertical position control is essential for sustaining high-performance operation in spherical tokamaks, where increased plasma elongation introduces stringent requirements on fast and robust stabilization. This work presents an experimentally validated reinforcement-learning(RL)-based vertical position control framework for the EXL-50U spherical tokamak. A high-fidelity discharge-reconstructed simulation environment is developed by integrating physics-based plasma-circuit models with experimental equilibrium information, enabling systematic controller synthesis and sim-to-real evaluation. Within this framework, RL is benchmarked in simulation against operational proportional--integral--derivative (PID) and model-based linear quadratic regulator (LQR) controllers under identical plant dynamics, actuator constraints, and measurement imperfections.Simulation results show that RL achieves tracking accuracy comparable to PID with consistently lower vertical-stabilization coil effort, while lightweight integral compensation improves robustness against residual model--plant mismatch. The RL controller is subsequently deployed on EXL-50U for closed-loop experiments. Across more than ten discharges with RL takeover, stable vertical regulation is achieved within the controlled windows. For seven representative discharges, RL maintains millimetre-scale tracking accuracy comparable to the operational PID controller (MAE typically ~ 1-5 mm) while consistently reducing actuator effort. These results demonstrate the feasibility of learning-based plasma control on a real spherical tokamak and establish a practical pathway toward future fusion control systems.

View source

Similar papers

Preprint Aug 2026

State-Space Model-Enabled Reinforcement Learning for Magnetic Configuration Controlon EXL-50U

Accurate feedback control of the plasma current ($I_p$) and centroid position $(R_c,Z_c)$ is essential for the stable operation of spherical torus (ST) plasmas. Conventional proportional-integral-derivative (PID) controllers require extensive manual tuning and struggle with the fast, strongly coupled dynamics that aris...

Peixin Guo, Zhengyuan Chen, Jianguo Chen et al. · 0 citations
Preprint Sep 2026

Observer-Based Model Predictive Control for Isoflux Regulation in the EXL-50U Spherical Tokamak

Isoflux control in spherical tokamaks requires coordinated plasma-current and boundary regulation under coil-voltage and computational constraints. This study develops observer-based constrained model predictive control (MPC) for EXL-50U. Vacuum-vessel spatial coarsening reduces a 538-state linearized model to 40 state...

Chang Lu, Jia-Yi Zhi, Xiang-Yang Guo et al. · 0 citations
Open access Nov 2026

Active Disturbance Rejection Control for Tilt-Rotor Aerial Manipulation Robots

While fully actuated tilt-rotor aerial robots offer enhanced dexterity over conventional systems, their complex architectures are susceptible to multi-band disturbances and unmodeled actuator dynamics. This work presents a comprehensive control framework to address these challenges. First, a novel frequency-selective n...

Hao-Yuan Liu, Xian-Gao Feng, Ming-Jing Wang et al. · 0 citations
Aug 2026

Synergistic learning and observation for robust attitude control of flexible spacecraft

Tomographic synthetic aperture radar (TomoSAR) missions require precise attitude control in the presence of low-frequency, weakly damped vibrations from large flexible appendages. To overcome the conservatism of model-based controllers and the lack of safety guarantees in data-driven approaches, this study aims to...

Yong Cao, Lijun Ye · 0 citations
Open access Sep 2026

Adaptive Neural PID Outer-Loop Control for Quadcopter UAVs with Asymmetric Saturation

An adaptive outer-loop neural PID control scheme with asymmetric saturation and clamping anti-windup for a quadcopter drone, using online training based on the Extended Kalman Filter, which mitigates the effects of actuator saturation and maintains operational feasibility under physical constraints.

Jose Olin Estrada, Jorge D. Rios, Alma Y. Alanis · 0 citations
Open access Sep 2026

Comparative Study of LQR and Neural Network Controllers for Quadcopter Roll Angle Stabilization

This study demonstrates how machine learning techniques can enhance classical control paradigms for aerial robotics and suggests that neural network controllers offer a viable path toward more adaptive and robust autonomous drone systems in challenging environments.

Hiba Arif, Marouane Kadi, Aymane Bidah et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.