This is the authors' abstract. We don't add key points for this paper.
Underground mines require continuous airflow monitoring, yet cabling or frequent battery replacement for distributed sensors is costly and hazardous. In this study, we present a piezoelectric wind energy harvester (PWEH) that also functions as a self-powered sensor for monitoring ventilation airflow conditions in underground mines. A shallow groove machined into the windward face of a D‑shaped bluff body boosts harvested power by 29.3% at a wind speed of 4.5 m/s compared with an ungrooved benchmark. The resulting voltage waveform supports two embedded sensing functions without external power: (i) ventilation fault diagnosis based on a stacking ensemble that integrates one-dimensional convolutional neural network (1D CNN), random forest (RF) and transfer-learned AlexNet classifier, enhanced by manifold‑based data augmentation, to identify 16 fault scenarios in a laboratory mine ventilation layout with up to 97.4% accuracy; and (ii) wind speed estimation using Random Forest that delivers a median mean absolute error below 0.12 m/s and worst‑case error under 0.5 m/s across 1.5–5.5 m/s airflow speeds, matching the performance of commercial anemometers. Laboratory trials in a scaled mine ventilation layout validate these capabilities across 16 duct networks with fan speeds ranging from 200–700 rpm, while energy measurements show the device produces a root mean square power output of 5.2 μW at a 4.5 m/s wind speed, which is sufficient to run a Bluetooth Low Energy-based SensorTag for fog data transmission every 50 seconds. By combining wind energy harvesting and airflow sensing in a single device, the proposed PWEH offers a battery-free and low-maintenance path to safer and smarter Mine Internet of Things deployments.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: