Advanced Materials & Sustainable Manufacturing 2024 , 1, 10003
12 of 14
4. Conclusions and Future Works At present, the papermaking industry lacks efficient means of whole life-cycle control. Therefore, this paper proposed a digital twin modelling framework for the paper manufacturing process, and developed two data completion methods for addressing the possible missing data in the modeling framework, including parameter solving based on mass and heat transfer mechanisms, and parameter prediction based on random forest. The following conclusions are obtained: the mechanism-based parameter solution can be used as a general method to solve a variety of parameter missing problems and usually obtains better results; the random forest-based parameter prediction model is robust and has high accuracy, with the average value of R 2 above 0.9. Base on which, this paper implemented a visual modelling of the surface condenser in the dry section of papermaking process based on CADSIM Plus and the digital twin framework. The model runs well and is able to monitor the dynamic change of parameters in real time. The digital twin-based visualization model proposed in this paper can reduce the coupling complexity of the physical entity modules, and has good scalability and generality, and can be subsequently extended to the whole process of paper production. However, there are certain limitations of the model should be further considered in the future study. Current studies mostly focused on the drying section, pay scarce attention to other sections, which is hard to integrate the processes as a whole. Meanwhile, certain simplified assumption is too ideal to be applied in the industry, which should be studied deeper in the future. In addition, artificial intelligence and big data analysis technology can be furthermore to be explored by online analysis on the basis of interaction with the production site thorough data. And applying the established models to implement management of production process for optimization and decision making, and so on. The integration of these techniques could significantly improve the production efficiency and reduce production costs, and ultimately achieve sustainable development of the process. Author Contributions Made substantial contributions to conception and design of the study and performed data analysis and interpretation: Z.L.; Performed data acquisition, as well as provided administrative, technical, and material support: J.L. and M.H. Ethics Statement Not applicable. Informed Consent Statement Not applicable. Funding This research received no external funding. Declaration of Competing Interest All authors declared that there are no conflicts of interest. References 1. Man Y, Han Y, Li J, Hong M. Review of energy consumption research for papermaking industry based on life cycle analysis. Chin. J. Chem. Eng. 2019 , 27 , 1543 – 1553. 2. Qian F, Bogle D, Wang M, Pistikopoulos S, Yan J. Artificial intelligence for smart energy systems in process industries. Appl. Energy 2022 , 324 , 119684. 3. Antonino PO, Capilla R, Pelliccione P, Schnicke F, Espen D, Kuhn T, et al. A Quality 4.0 Model for architecting industry 4.0 systems. Adv. Eng. Inform. 2022 , 54 , 101801. 4. Fan Y, Dai C, Huang S, Hu P, Wang X, Yan M. A life-cycle digital-twin collaboration framework based on the industrial internet identification and resolution. Int. J. Adv. Manuf. Technol. 2022 , 123 , 2883 – 2911. 5. Hu Y, Li J, Hong M, Ren J, Man Y. Industrial artificial intelligence based energy management system: Integrated framework for electricity load forecasting and fault prediction. Energy 2022 , 244, 123195.
Made with FlippingBook interactive PDF creator