The Digital Twin Model of Chemical Production Systems in Smart Factories: A Case Study.

Nowadays, more and more information technology such as internet-of-things (IoT) and artificial intelligence (AI) are applied to all walks of life. Facing the pressure of technological innovation, traditional manufacturers have paid attention to digital twin to improve production lines. However, most of the research on digital twin have not implement the concept of digital twin into actual production lines. To fill this gap, this paper demonstrates a digital twin application case study with whole process analysis based on machine learning. More specifically, data preprocessing, clustering, classification and regression are utilized to build a digital twin model. Moreover, we deploy digital twin model to cloud server and develop web application with real-time diagnosis function which has user-friendly interactive interface. Furthermore, the digital twin model of chemical production system has already been implemented in a physical chemical process plant. So the results of the experiment can prove the effectiveness of our proposed model building approaches.