پایش وضعیت دیزل‌ژنراتور با ترکیب داده‌های ارتعاش، صدا و حسگرهای محیطی مبتنی بر مدل‌های یادگیری ماشین

Diesel generators play an important role in providing stable power supply in critical infrastructures such as hospitals and data centers. Any failure or degradation of these equipment can lead to process shutdown and heavy repair costs. In this study, an intelligent approach for diesel generator fault diagnosis is presented, which is based on the analysis of multi-sensor data including vibration, noise, and environmental and operational parameters (engine temperature, pressure, humidity, altitude, oxygen). In the case study, the effect of air filter clogging was created at several different levels and the relevant data was collected by the designed portable device. After preprocessing and feature extraction, several machine learning models including SVM, KNN, random forest, and artificial neural network were evaluated for status classification. The results showed that the neural network had the best performance and was able to classify the air filter clogging status with 100% accuracy. The findings of this research show that combining multiple sensors with machine learning algorithms is an effective tool for early fault detection and implementation of predictive maintenance in diesel generators, and can prevent sudden stops and repair costs.