An Anomaly Detection Method for Conveyor System Based on Digital Twin

With the continuous evolution of the Internet of Things (IoT), it drives the ongoing advancement of industry. The widespread application of IoT enables factories to achieve intelligent production management, thus enhancing factory production efficiency. With complex factory equipment and expanding production scale, diverse anomalies are challenging to identify using traditional methods. To address these issues, digital twin technology can realize real-time monitoring of production to help managers deal with anomalies promptly, thus improving the accuracy of anomaly detection. Taking the factory conveyor equipment as the research object, a factory-oriented anomaly detection method based on digital twin is proposed to improve the productivity and safety of the conveyor. Firstly, we propose the design of a digital twin module for anomaly detection. This module facilitates the interaction and visualization of data between the physical entity and virtual system, thereby guiding the normal operation of the production process. Secondly, we have developed an anomaly detection method to tackle the issue of class imbalance and explore the correlation of time-series data. The experimental results demonstrate that the anomaly detection method, driven by the digital twin, effectively identifies equipment anomalies.