Logo image
Digital Twin Model Development for Chemical Plants Using Multiple Time-Steps Prediction Data-Driven Model and Rolling Training
Book chapter   Peer reviewed

Digital Twin Model Development for Chemical Plants Using Multiple Time-Steps Prediction Data-Driven Model and Rolling Training

Jia-Lin Kang, Somayeh Mirzaei, Yao-Chen Lee, Yao-Cheng Chuang, Marvin Frias, Cheng-Huang Chou, San-Jang Wang, David Shan Hill Wong and Shi-Shang Jang
Computer Aided Chemical Engineering, Vol.50, pp.567-572
01/2021

Abstract

Digital Twin long-term prediction rolling training sequence-to-sequence model Chemical Engineering (all) Computer Science Applications
Data-driven operation monitoring and optimization of chemical plants can be performed using Digital Twin, along with intelligent algorithms. This study proposed a sequence-to-sequence rolling training algorithm to overcome the challenge of rolling predictions. The data were generated through dynamic simulation of the vapor-recompression C3 process using Aspen Plus. Studies showed that StS with rolling training could better fit the real data than the StS model. Moreover, StS with rolling training was able to present efficient long-term predictions as Digital Twin.

Metrics

1 Record Views

Details

Logo image