DEVELOPMENT AND PERFORMANCE EVALUATION OF AN ENERGY-EFFICIENT SMART MONITORING SYSTEM FOR INDUSTRIAL EQUIPMENT

Main Article Content

Johnathan Miller Greenfield International Academy, USA

Abstract

The increasing adoption of Industry 4.0 technologies has created a growing demand for intelligent, low-power, and cost-effective systems capable of continuously monitoring industrial equipment. Conventional condition-monitoring solutions often depend on wired infrastructure, high-power data acquisition units, and centralized processing platforms, which may limit their applicability in geographically distributed or difficult-to-access industrial environments. This study proposes an energy-efficient smart monitoring system designed to observe the operating condition of industrial equipment through vibration, temperature, current, and rotational-speed measurements. The developed architecture combines low-power sensors, an embedded microcontroller, local signal processing, wireless communication, and a cloud-based monitoring dashboard. A duty-cycling strategy, adaptive sampling algorithm, threshold-based anomaly detection, and event-triggered communication mechanism were incorporated to minimize energy consumption without substantially reducing monitoring reliability. System performance was evaluated through controlled laboratory experiments conducted under normal operation, mechanical imbalance, bearing degradation, overheating, and overload conditions. The prototype achieved an illustrative anomaly-detection accuracy of 94.6%, an average wireless packet-delivery rate of 97.8%, and an alert latency of 1.42 s. Adaptive sampling and event-based transmission reduced average energy consumption by approximately 48.3% compared with continuous sensing and periodic communication. The findings indicate that combining edge processing with low-power wireless communication can provide reliable and scalable condition monitoring for motors, pumps, fans, compressors, and other rotating industrial equipment. The proposed approach can support predictive maintenance, reduce unplanned downtime, and improve the energy efficiency of industrial monitoring infrastructures.


 

Article Details

Section
Articles

References

Banks, A., Briggs, E., Borgendale, K., & Gupta, R. (Eds.). (2019). MQTT Version 5.0. OASIS Standard.

LoRa Alliance. (2023). LoRaWAN Specification Version 1.1.

LoRa Alliance. (2023). LoRa Alliance technical stack: Low-power wide-area networking for battery-operated IoT devices.

de la Fuente, R., Radrigan, L., & Morales, A. S. (2024). Enhancing predictive maintenance in mining mobile machinery through a TinyML-enabled hierarchical inference network. arXiv.

Prajwal, B. N., Yelchuri, H., Shastry, V., & Prabhakar, T. V. (2022). PreMa: Predictive maintenance of solenoid valve in real time at embedded edge level. arXiv.

Moin, A., Challenger, M., Badii, A., & Günnemann, S. (2021). Supporting AI engineering on the IoT edge through model-driven TinyML. arXiv.

Langer, T., Widra, M., & Beyer, V. (2025). TinyML towards Industry 4.0: Resource-efficient process monitoring of a milling machine. arXiv.