Abstract: To address the challenge of safety risk early warning in petrochemical plants, this study develops a dynamic early warning model based on digital twin technology. It first sorts out the risk characteristics and technical foundations, constructs a full-process early warning system of "perception-modeling-analysis-early warning-management and control", and designs a multi-source data collection and fusion scheme. Then, a multi-scale digital twin model is established, and a dynamic risk assessment model integrating LSTM and GCN is constructed. Experimental verification on an ethylene cracking unit shows the model achieves an early warning accuracy of 94.2% with an average advance warning time of 37 minutes, effectively enhancing the initiative and accuracy of safety prevention and control. This research provides theoretical and practical support for risk management in the digital transformation of the petrochemical industry.
Gao et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: