Answers are grounded in this paper and cite their sources.
Predictive maintenance relies on Internet of Things (IoT) measurements to estimate the condition of critical equipment via data-driven schemes. In this paper, an IoT system for fault diagnosis of electrical-mechanical equipment is presented consisting of a smart node and a web application for early warning. The smart node acquires current time-series, and the obtained signals are analyzed utilizing a hybrid method that combines advanced signal processing techniques and Deep Learning models. More specifically, the current time- series data undergoes Discrete Wavelet Transform (DWT) analysis to capture the harmonic frequencies associated with faults, allowing for the extraction of time-frequency information. Subsequently, the extracted features are fed into a (Convolutional Neural Networks) CNN model to capture spatial information. The motivation for implementing the proposed IoT system for fault diagnosis is that there is a need for automated condition monitoring of equipment since traditional manual monitoring methods are often time- consuming, labor-intensive, and prone to human errors. Additionally, the proposed IoT system can diagnose indistinguishable and complex faults that human operators could not.
No takes yet. Share an insight, caveat, or question.
Paraskevopoulos et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: