The 2025 PNPL competition presents over 50 hours of within-subject MEG data, defines speech detection and phoneme classification benchmarks with F1-macro scoring, and reports reference baselines of 68.04% and 60.39%.
Post-corona unipolar chargers with tuneable aerosol size-charge relations: Parameters affecting ion dispersion and particle trajectories for charger designs
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
This paper focusses on the mean charge per particle of monodisperse submicron aerosol, charged by diffusion of unipolar ions in post-corona discharge. It aims to confirm and discuss the limits of considering a single value of Ni$\times$t to describe aerosol charging and then to present methods to control the size-charge relation. Three aerosol chargers, with different mixings of ion and aerosol flows are investigated. Despite comparable ion sources with discharge currents of a few tens of $\mu$A, the size-charge relations differs from one charger to another due to different ion-aerosol mixing conditions and subsequent ion density along particles trajectories. Discrepancies are even more noticeable as the particle size increases. Discharge current, velocities of ion and aerosol flows and electric field control post-discharge ion density in each point of the charging volume. The control of particle trajectory in expanding unipolar ion cloud, leads to tuneable size-charge relations. Aerosol inertia and charging dynamics, that both depends on particle size, affects the Ni$\times$t experienced by the particle and thus the final charge of the particle. Operating conditions to reach a constant mean charge for particles larger than 200 nm are reported. Conclusions provide a basis to design aerosol chargers devoted to electric mobility selection for aerosol deposition, separation or electrical measurements especially to overcome the limits of mobility-to-size data inversion due to multiple charge ambiguity using diffusion chargers.
citation-role summary
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
The 2025 PNPL Competition: Speech Detection and Phoneme Classification in the LibriBrain Dataset
The 2025 PNPL competition presents over 50 hours of within-subject MEG data, defines speech detection and phoneme classification benchmarks with F1-macro scoring, and reports reference baselines of 68.04% and 60.39%.