A carregar...
2 resultados
Resultados da pesquisa
A mostrar 1 - 2 de 2
- Non-intrusive low-cost IoT-based hardware system for sustainable predictive maintenance of industrial pump systemsPublication . Brito, Sergio; Azinheira, Gonçalo; Semião, Jorge; Sousa, Nelson; Pérez Litrán, SalvadorIndustrial maintenance has shifted from reactive repairs and calendar-based servicing toward data-driven predictive strategies. This paper presents a non-intrusive, low-cost IoT hardware platform for sustainable predictive maintenance of rotating machinery. The system integrates an ESP32-S3 sensor node that captures vibration (100 kHz) and temperature data, performs local logging, and communicates wirelessly. An automated spectral band segmentation framework is introduced, comparing equal-energy, linear-width, nonlinear, clustering, and peak-valley partitioning methods, followed by a weighted feature scheme that emphasizes high-value bands. Three unsupervised one-class classifiers-transformer autoencoders, GANomaly, and Isolation Forest-are evaluated on these weighted spectral features. Experiments conducted on a custom pump test bench with controlled anomaly severities demonstrate strong anomaly classification performance across multiple configurations, supported by detailed threshold-characterization metrics.
- A cost-effective solution for predictive maintenance in industrial pumping systemsPublication . Brito, Sergio; Azinheira, Gonçalo; Semião, Jorge; Sousa, NelsonThis paper presents a cost-effective, Internet of Things (IoT)-based solution for predictive maintenance (PdM) in industrial pumping systems. The proposed system integrates custom-built hardware with machine learning (ML) algorithms to monitor and detect anomalies in real-time. The innovation of the system lies in its non-intrusive design, ease of installation, and adaptability to a variety of industrial environments, providing a practical, low-cost alternative to traditional PdM solutions. Detailed discussion is provided on the hardware component selection, which prioritizes affordability without sacrificing performance, as well as the machine learning strategies used for anomaly detection. Preliminary results from laboratory and field testing demonstrate the system’s potential for reducing downtime and maintenance costs, with a focus on extending the application to broader industrial contexts.