Related papers: Deep or Not Deep: Supervised Learning Approaches t…
Wide accessibility of imaging and profile sensors in modern industrial systems created an abundance of high-dimensional sensing variables. This led to a a growing interest in the research of high-dimensional process monitoring. However,…
The electron density is a key parameter to characterize any plasma. Most of the plasma applications and research in the area of low-temperature plasmas (LTPs) are based on the accurate estimations of plasma density and plasma temperature.…
Controlling and monitoring plasma within a tokamak device is complex and challenging. Plasma off-normal events, such as disruptions, are hindering steady-state operation. For large devices, they can even endanger the machine's integrity and…
Model-based plasma scenario development lies at the heart of the design and operation of future fusion powerplants. Including turbulent transport in integrated models is essential for delivering a successful roadmap towards operation of…
The embedding-based representation learning is commonly used in deep learning recommendation models to map the raw sparse features to dense vectors. The traditional embedding manner that assigns a uniform size to all features has two…
Deep learning is a powerful tool to represent subgrid processes in climate models, but many application cases have so far used idealized settings and deterministic approaches. Here, we develop stochastic parameterizations with calibrated…
Edge-Localized-Mode (ELM) suppression by resonant magnetic perturbations (RMPs) generally occurs over very narrow ranges of the plasma current (or magnetic safety factor q95) in the DIII-D tokamak. However, wide q95 ranges of ELM…
Energetic particle (EP) driven instabilities will be of strongly increased relevance in future burning plasmas as the EP pressure will be very large compared to the thermal plasma. Understanding the interaction of EPs and bulk plasma is…
The nonlinear dynamics of energetic particle (EP) driven geodesic acoustic modes (EGAM) in tokamaks is investigated, and compared with the beam-plasma system (BPS). The EGAM is studied with the global gyrokinetic (GK) particle-in-cell code…
The peak pressure gradient in the pedestal (dPe/d{\psi}) on MAST varies little between edge localized modes (ELMs), although it varies between discharges due factors such as gas fuelling and plasma current. The pressure pedestal width in…
The rise of deep learning has marked significant progress in fields such as computer vision, natural language processing, and medical imaging, primarily through the adaptation of pre-trained models for specific tasks. Traditional…
The development of small- and no-ELM regimes for ITER is a high priority topic due to the risks associated to type-I ELMs. By considering non-linear extended MHD simulations of the ASDEX Upgrade tokamak with the JOREK code, we probe a…
In some conditions, I-mode plasmas can feature pedestal relaxation events (PREs) that transiently enhance the energy reaching the divertor target plates. To shed light into their appearance, characteristics and energy reaching the divertor…
Machine learning for scientific applications faces the challenge of limited data. We propose a framework that leverages a priori known physics to reduce overfitting when training on relatively small datasets. A deep neural network is…
Deep Learning methods are known to suffer from calibration issues: they typically produce over-confident estimates. These problems are exacerbated in the low data regime. Although the calibration of probabilistic models is well studied,…
We present a deep learning emulator for stochastic and chaotic spatio-temporal systems, explicitly conditioned on the parameter values of the underlying partial differential equations (PDEs). Our approach involves pre-training the model on…
Standard approaches to neoclassical theory do not extend into regions of strong gradients in tokamaks such as the pedestal and internal transport barriers. Here, we calculate the modifications to neoclassical electron physics inside strong…
We present TokaMind, an open-source foundation model framework for fusion plasma modeling, based on a Multi-Modal Transformer (MMT) and trained on heterogeneous tokamak diagnostics from the publicly available MAST dataset. TokaMind supports…
Error detection (ED), which aims to identify incorrect or inconsistent cell values in tabular data, is important for ensuring data quality. Recent state-of-the-art ED methods leverage the pre-trained knowledge and semantic capability…
An accurate evolution model is crucial for effective control and in-depth study of fusion plasmas. Evolution methods based on physical models often encounter challenges such as insufficient robustness or excessive computational costs. Given…