An Internet-of-Things-Integrated Deep Learning Model for Fault Diagnosis in Industrial Rotating Machines

Rotating machines are crucial in industries for reliable system operation, but unexpected failures can result in significant financial losses and personnel damage. Hence, fault diagnosis is important. Among the common types of faults in rotating machines, imbalance, and misalignment are important types of faults in operation but are rarely studied. This research utilizes data from four distinct operating conditions: normal operation, imbalanced condition, imbalanced condition combined with horizontal misalignment, and imbalanced condition combined with vertical misalignment. This study proposed a method for fault diagnosis using Continuous Wavelet Transform (CWT) for extracting vibration signals into RGB (Red – Green - Blue) images and using Convolutional Neural Network (CNN) models for fault diagnosis. Additionally, Internet of Things (IoT) integration is implemented using the Message Queuing Telemetry Transport (MQTT) protocol, enabling the update of fault status to the cloud for efficient monitoring. A mobile application provides an intuitive platform for visualizing and interacting with fault data. Moreover, several machine learning (ML) models are also utilized for comparison with the proposed method such as Random Forest (RF), Multilayer Perceptron (MLP), and Xtreme Gradient Boosting (XGBoost). The results indicate that CNN yields an overall accuracy of 93.27% compared to other ML methods.