Development of a Vibration Monitoring System for Predictive Maintenance of Industrial Generators in Anambra State

Industrial generators are critical power assets for manufacturing facilities in Anambra State, South East Nigeria, where grid power supply is unreliable. Unplanned generator failures due to mechanical faults cause costly production downtime, yet vibration-based predictive maintenance systems are rarely deployed at the small enterprise level. This project presents the development of a vibration monitoring system for detecting early-stage mechanical anomalies in industrial diesel generators. An ADXL345 three-axis accelerometer is mounted on the generator casing and interfaced with an Arduino Mega microcontroller that samples vibration data at 200 Hz. Fast Fourier Transform analysis of the time-domain vibration signal is performed to extract frequency-domain features indicative of bearing wear, imbalance, and misalignment. Baseline vibration signatures were established for a 45 kVA Perkins diesel generator under various load conditions, and alert thresholds were defined for amplitude and frequency deviation. The system was deployed at a bottling plant in Onitsha for three months. During this period, the system correctly identified a developing bearing fault 11 days before it caused a generator shutdown, enabling preventive maintenance intervention. Data is displayed on a local LCD and transmitted via Wi-Fi to a monitoring laptop. Component cost for the sensing and processing unit was approximately 28,000 Naira. The study recommends integration with a cloud-based asset management platform for multi-generator facility deployments. Keywords: vibration monitoring, predictive maintenance, generator, FFT analysis, Anambra State