IoT-Based Real-Time Vibration and Temperature Monitoring System for Industrial Machinery Using ESP32 and MQTT

Abstract

This study presents the design and validation of an Internet of Things (IoT)–based real-time vibration and temperature monitoring system for industrial machinery using an ESP32 microcontroller and MQTT communication. The proposed system addresses limitations of periodic manual inspection by enabling continuous monitoring with on-device signal processing and direct compliance evaluation with ISO 10816-3. The main contribution of this work is the implementation of ISO-based vibration severity classification directly at the edge level, integrating multi-sensor acquisition with real-time Root Mean Square (RMS) and Fast Fourier Transform (FFT) processing without relying on predictive or machine learning algorithms. This architecture enables low-latency decision support, reduced bandwidth usage, and improved system independence from cloud computation. The system integrates two ADXL345 vibration sensors and two temperature sensors into a single ESP32 node for synchronized monitoring. Experimental validation on an industrial reciprocating compressor demonstrated stable data acquisition and 100% communication availability during testing. RMS vibration values ranged from 2.15 to 2.17 mm/s, with operating temperatures around 67 °C. FFT analysis identified dominant frequencies consistent with machine characteristics. According to ISO 10816-3 classification, the monitored condition was within safe to early warning levels, confirming the reliability and practical feasibility of the proposed edge-based monitoring approach for condition-based maintenance.