See how cycle-by-cycle numerical analysis of current waveforms uses composite spike detection a
Demodulating PV current removes DC and periodic noise, while autocorrelation energy thresholds cut false alarms and catch intermittent arc faults.
Signal envelopes and DBSCAN isolate power system event outliers in massive synchrophasor data, enabling faster detection and ML classification.
A staged XLPE formulation workflow balances crosslinking, antioxidants, and by-product control to improve cable insulation properties.
Centralized high-frequency arc analysis cuts branch AFCI complexity and cost while enabling coordinated fault recovery attempts.
Neural-network arc detection improves real-time DC fault identification in photovoltaic circuits, reducing misidentification and equipment damage.
Odd-order harmonic and phase-angle analysis separates partial discharge from motor operating noise for earlier insulation fault detection.
Physics-based modeling predicts pressure rise and volatile generation at composite fastener interfaces to assess lightning ignition risk early.
Distributed conductive slugs and plungers spread high voltage in semiconductor package testing to prevent arcing and dielectric failure.
A switched test circuit isolates a capacitor, applies a test voltage, and checks measured thresholds to detect open connections in high-voltage chargers.
Real-time ripple and ageing-curve modelling predicts DC link capacitor remaining life before capacitance and ESR drift destabilize the converter.
Neural-network arc detection with Fourier analysis and normalization improves real-time DC fault accuracy in photovoltaic circuits.
A dual three-phase PMSM doubles as a charging transformer and traction motor, cutting EV charger-inverter weight, cost, and redundancy.
A shared magnetic core monitors combined currents from multiple power lines, enabling arc detection without separate sensors on each line.