Search Results (2)

The challenge for ultrasonic (US) power transfer systems, in implanted/wearable medical devices, is to determine when misalignment occurs (e.g., due to body motion) and apply directional correction accordingly. In this study, a number of machine learning algorithms were evaluated to classify US transducer misalignment, based on data signal transmissions between the transmitter and receiver. Over seven hundred US signals were acquired across a range of transducer misalignments. Signal envelopes and spectrograms were used to train and evaluate machine learning (ML) algorithms, classifying misalignment extent. The algorithms included an autoencoder, convolutional neural network (CNN) and neural network (NN). The best performing algorithm, was deployed onto a TinyML device for evaluation. Such systems exploit low power microcontrollers developed specifically around edge device applications, where algorithms were configured to run on low power, restricted memory systems. TensorFlow Lite and Edge Impulse, were used to deploy trained models onto the edge device, to classify signals according to transducer misalignment extent. TinyML deployment, demonstrated near real-time (<350 ms) signal classification achieving accuracies > 99%. This opens the possibility to apply such ML alignment algorithms to US arrays (capacitive micro-machined ultrasonic transducer (CMUT), piezoelectric micro-machined ultrasonic transducer (PMUT) devices) capable of beam-steering, significantly enhancing power delivery in implanted and body worn systems. Full article

FNS-ST is a fusion neutron source project based on a spherical tokamak (R/a = 0.5 m/0.3 m) with a steady-state neutron generation of ~10¹⁸ n/s. Neutral beam injection (NBI) is supposed to maintain steady-state operation, non-inductive current drive and neutron production in FNS-ST plasma. In a low aspect ratio device, the toroidal magnetic field shape is not optimal for fast ions confinement in plasma, and the toroidal effects are more pronounced compared to the conventional tokamak design (with R/a > 2.5). The neutral beam production and the tokamak plasma response to NBI were efficiently modeled by a specialized beam-plasma software package BTR-BTOR, which allowed fast optimization of the neutral beam transport and evolution within the injector unit, as well as the parametric study of NBI induced effects in plasma. The “Lite neutral beam model” (LNB) implements a statistical beam description in 6-dimensional phase space (10⁶–10¹⁰ particles), and the beam particle conversions are organized as a data flow pipeline. This parametric study of FNS-ST tokamak is focused on the beam-plasma coupling issue. The main result of the study is a method to achieve steady-state current drive and fusion controllability in beam-driven toroidal plasmas. LNB methods can be also applied to NBI design for conventional tokamaks. Full article