A Cloud-Edge Collaborative Framework for Industrial IoT Device Fault Detection and Prediction

This paper proposes a cloud-edge collaborative framework for industrial IoT device fault detection and prediction. The framework leverages the real-time processing and lowlatency characteristics of edge computing along with the powerful computational and storage capabilities of cloud computing to achieve efficient monitoring and fault prediction for industrial equipment. At the edge layer, lightweight algorithms are deployed for real-time data collection and preliminary fault detection; at the cloud layer, deep learning models are utilized for historical data analysis and fault trend prediction. Experimental results demonstrate that compared to traditional single computing paradigms, our framework shows significant improvements in detection accuracy, prediction time window, and system response time, effectively enhancing industrial production efficiency and equipment reliability.