With the advent of the fourth technological revolution,the new generation of artificial intelligence(AI)has imparted new significance and opportunities to the modeling of momentum,heat,and mass transfer,as well as che...With the advent of the fourth technological revolution,the new generation of artificial intelligence(AI)has imparted new significance and opportunities to the modeling of momentum,heat,and mass transfer,as well as chemical reaction processes with the realm of chemical engineering.AI techniques are being widely employed in the chemical industry and are constantly evolving to offer more effective solutions for tackling practical challenges.This review delves the transformation of the chemical industry from traditional digital simulations to advanced AI-based approaches,targeting high efficiency and low carbon emissions across the scale from molecules to factories.Particular emphasis is mainly placed on the research carried out within the research group of Weifeng Shen.At the molecular level,the intelligent capture of molecular characteristics and the precise determination of structure-property relationships have reached a mature stage.Furthermore,multifunction-driven reverse molecular design for solvents,reaction reagents,and other substances has been accomplished through AI-based high-throughput screening and generative models.To improve the safety,environmental friendliness,and carbon reduction performance of chemical separation processes,a series of innovative reinforcement strategies have been put forward,with a primary focus on the systematic optimization of solvent design.On the process scale of actual production,it frequently occurs that the constructed mechanism model fails to align with the actual system behavior,thereby restricting the industrial application of the model.To solve this issue,mechanism-data hybrid-driven frameworks have been successfully developed,leveraging AI-enhanced prediction,diagnosis,optimization,and control for complex separation systems in practice.Finally,as a bridge connecting big data intelligent technology and actual industrial processes,dynamic digital twin modeling is discussed for its potential to boost efficiency and sustainability in the chemical industry.展开更多
基金support provided by the National Natural Science Foundation of China(Grant No.22278044)the Chongqing Science Fund for Distinguished Young Scholars(Grant No.CSTB2022NSCQ-JQX0021)+5 种基金the Fundamental Research Funds for the Central Universities(Grant No.2024IAIS-QN004)the Chongqing Innovation Support Key Program for Returned Overseas Chinese Scholars(Grant No.CX2023002)the Key Project of Technical Innovation and Application Development(Grant No.CSTB2024TIAD-KPX0058)the Science and Technology Innovation Key R&D Program of Chongqing(Grant No.CSTB2024TIAD-STX0032)the Xinjiang Autonomous Region Regional Collaborative Innovation Special Science and Technology Assistance Plan Project(Grant No.2024E02036)the Open Research Project of the State Key Laboratory of Industrial Control Technology,China(Grant No.ICT2024B01).
文摘With the advent of the fourth technological revolution,the new generation of artificial intelligence(AI)has imparted new significance and opportunities to the modeling of momentum,heat,and mass transfer,as well as chemical reaction processes with the realm of chemical engineering.AI techniques are being widely employed in the chemical industry and are constantly evolving to offer more effective solutions for tackling practical challenges.This review delves the transformation of the chemical industry from traditional digital simulations to advanced AI-based approaches,targeting high efficiency and low carbon emissions across the scale from molecules to factories.Particular emphasis is mainly placed on the research carried out within the research group of Weifeng Shen.At the molecular level,the intelligent capture of molecular characteristics and the precise determination of structure-property relationships have reached a mature stage.Furthermore,multifunction-driven reverse molecular design for solvents,reaction reagents,and other substances has been accomplished through AI-based high-throughput screening and generative models.To improve the safety,environmental friendliness,and carbon reduction performance of chemical separation processes,a series of innovative reinforcement strategies have been put forward,with a primary focus on the systematic optimization of solvent design.On the process scale of actual production,it frequently occurs that the constructed mechanism model fails to align with the actual system behavior,thereby restricting the industrial application of the model.To solve this issue,mechanism-data hybrid-driven frameworks have been successfully developed,leveraging AI-enhanced prediction,diagnosis,optimization,and control for complex separation systems in practice.Finally,as a bridge connecting big data intelligent technology and actual industrial processes,dynamic digital twin modeling is discussed for its potential to boost efficiency and sustainability in the chemical industry.
