Li-Tao Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Li-Tao Zhu has a total of 110 co-authors. Here are the top co-authors by number of shared publications:
- Zheng-Hong Luo
Ningxia University
26 shared publications Zheng-Hong Luo
11 shared publications- Bo Ouyang
Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites Shanghai Jiao Tong University Shanghai People's Republic of China
5 shared publications Ling Qin
5 shared publications- He Lei
State Key Laboratory of Polyolefins and Catalysis, Shanghai Key Laboratory of Catalysis Technology for Polyolefins, School of Chemistry and Chemical Engineering
4 shared publications - Hui Pan
Shanghai Key Laboratory of Materials Protection and Advanced Materials in Electric Power, Shanghai University of Electric power, Shanghai 200090, China
4 shared publications
Latest Publications
- 3 October 2026
- Published by American Chemical Society (ACS) in Energy & Fuels
- Vol. 40 (41) , 22454-22472
Physics-Guided Modeling and Artificial Intelligence for Proton Exchange Membrane Fuel Cells in Extreme Environments
Proton exchange membrane fuel cells (PEMFCs) deployed in aviation, underwater, and space applications encounter different combinations of environmental and mission constraints. These include altitude-dependent oxygen supply, flight-load transients, closed reactant loops, constrained heat rejection, and gravity-dependent two-phase transport. The resulting changes in reactant supply, water–thermal balance, and degradation complicate system integration. Intelligent methods can support rapid modeling, state estimation, health management, and constrained control, yet their applicability depends on the operating environment and validation evidence. This review connects environment-specific mechanisms with modeling and deployment requirements, distinguishing three evidence levels: direct studies under representative extreme conditions, studies reproducing individual relevant stressors, and transferable methods established under conventional PEMFC conditions. The synthesis identifies gaps in data representativeness, cross-domain generalization, physical consistency, uncertainty assessment, and closed-loop validation. An integrated physics–data–decision framework is then proposed to connect environment-aware modeling, state and health estimation, risk-aware control, and staged validation. Its purpose is to guide method selection and identify the evidence needed for reliable onboard operation in aerial, underwater, and space systems.
- 3 October 2026
- Published by American Chemical Society (ACS) in Energy & Fuels
- Vol. 40 (41) , 22454-22472
Physics-Guided Modeling and Artificial Intelligence for Proton Exchange Membrane Fuel Cells in Extreme Environments
Proton exchange membrane fuel cells (PEMFCs) deployed in aviation, underwater, and space applications encounter different combinations of environmental and mission constraints. These include altitude-dependent oxygen supply, flight-load transients, closed reactant loops, constrained heat rejection, and gravity-dependent two-phase transport. The resulting changes in reactant supply, water–thermal balance, and degradation complicate system integration. Intelligent methods can support rapid modeling, state estimation, health management, and constrained control, yet their applicability depends on the operating environment and validation evidence. This review connects environment-specific mechanisms with modeling and deployment requirements, distinguishing three evidence levels: direct studies under representative extreme conditions, studies reproducing individual relevant stressors, and transferable methods established under conventional PEMFC conditions. The synthesis identifies gaps in data representativeness, cross-domain generalization, physical consistency, uncertainty assessment, and closed-loop validation. An integrated physics–data–decision framework is then proposed to connect environment-aware modeling, state and health estimation, risk-aware control, and staged validation. Its purpose is to guide method selection and identify the evidence needed for reliable onboard operation in aerial, underwater, and space systems.
- 31 August 2026
- Published by Springer Nature in ENGINEERING Energy
- Vol. 20 (5) , 10958
Artificial intelligence for plastic pyrolysis to produce fuels and chemicals: From reaction kinetics and reactor design to intelligent agent development
- 31 August 2026
- Published by Springer Nature in ENGINEERING Energy
- Vol. 20 (5) , 10958
Artificial intelligence for plastic pyrolysis to produce fuels and chemicals: From reaction kinetics and reactor design to intelligent agent development
- Open Access
Multiscale Modelling of Thermal Runaway in Lithium-Ion Batteries
Thermal runaway of lithium-ion batteries involves rapid heat release, gas generation, and multiphase transport, but their interaction inside a cell remains difficult to resolve. A multiscale computational fluid dynamics model was developed for a single 18650 cell by coupling microscale reaction kinetics, mesoscale interfacial heat transfer, and macroscale gas–liquid transport with a stationary porous-solid energy balance. The model describes internal temperature and the evolution of carbon dioxide, oxygen, water vapour, and hydrogen fluoride while examining the effects of porosity and the modelled dimethyl carbonate mass fraction. The medium-to-fine grid difference in carbon dioxide mass fraction was approximately 0.16%. Time steps of 0.01, 0.001, and 0.0001 s produced mass fractions of 0.0564, 0.0617, and 0.0618, respectively. Increasing the solvent mass fraction and porosity primarily shortened the induction period, while the peak temperature and terminal species levels remained similar. A quadratic response surface fitted to the simulation database was searched using grey wolf, genetic, and particle swarm methods. Grey wolf and particle swarm gave candidate times to peak temperature of about 238.8 s, whereas the genetic method gave 237.2 s, a difference of 1.6 s (0.67%). Particle swarm reached the high-response region within fewer iterations, while grey wolf maintained broader exploration. The proposed model connects reaction kinetics with macroscopic temperature and species evolution and clarifies how electrolyte composition and porous structure regulate the time scale of thermal runaway.
Open AccessMultiscale Modelling of Thermal Runaway in Lithium-Ion Batteries
Thermal runaway of lithium-ion batteries involves rapid heat release, gas generation, and multiphase transport, but their interaction inside a cell remains difficult to resolve. A multiscale computational fluid dynamics model was developed for a single 18650 cell by coupling microscale reaction kinetics, mesoscale interfacial heat transfer, and macroscale gas–liquid transport with a stationary porous-solid energy balance. The model describes internal temperature and the evolution of carbon dioxide, oxygen, water vapour, and hydrogen fluoride while examining the effects of porosity and the modelled dimethyl carbonate mass fraction. The medium-to-fine grid difference in carbon dioxide mass fraction was approximately 0.16%. Time steps of 0.01, 0.001, and 0.0001 s produced mass fractions of 0.0564, 0.0617, and 0.0618, respectively. Increasing the solvent mass fraction and porosity primarily shortened the induction period, while the peak temperature and terminal species levels remained similar. A quadratic response surface fitted to the simulation database was searched using grey wolf, genetic, and particle swarm methods. Grey wolf and particle swarm gave candidate times to peak temperature of about 238.8 s, whereas the genetic method gave 237.2 s, a difference of 1.6 s (0.67%). Particle swarm reached the high-response region within fewer iterations, while grey wolf maintained broader exploration. The proposed model connects reaction kinetics with macroscopic temperature and species evolution and clarifies how electrolyte composition and porous structure regulate the time scale of thermal runaway.
- 5 May 2026
- Published by American Chemical Society (ACS) in Industrial & Engineering Chemistry Research
- Vol. 65 (19) , 9960-9969
Kinetic Study and Microreactor Modeling for the Polycondensation of p-Phenylenediamine and Terephthaloyl Chloride
- 5 May 2026
- Published by American Chemical Society (ACS) in Industrial & Engineering Chemistry Research
- Vol. 65 (19) , 9960-9969
Kinetic Study and Microreactor Modeling for the Polycondensation of p-Phenylenediamine and Terephthaloyl Chloride
- Open Access
- 19 March 2026
- Published by Scilight Press Pty Ltd in Smart Chemical Engineering
- Vol. 2 (1) , 3
Large Language Models for Automating Computational Fluid Dynamics (CFD): From Predictive Modeling and Optimization to Execution Scheduling
Artificial intelligence (AI), especially large language models (LLMs), are triggering an intelligent transformation in the field of computational fluid dynamics (CFD). LLMs can assist and even partially replace humans in executing complex fluid simulation tasks. Due to such opportunities, we systematically review recent progress and limitations of applying LLMs to the CFD. Specifically, three key applications of LLMs in CFD are surveyed, explored, and analyzed, including: (1) Predictive modeling for prediction of fluid behavior (such as real-time generation of flow field distributions) and automatic discovery of turbulence models (such as improvement of the k-ε model based on direct numerical simulation data, achieving implicit encoding of physical laws); (2) Optimization applications, including CFD model optimization (i.e., optimization of numerical model parameters and their sub-models, such as turbulence model parameters), hyperparameter optimization of machine learning models (such as learning rate, optimizer, number of neurons, and hidden layers involved in neural networks training), geometric structure optimization (such as topological optimization of fluid process equipment structures), and process operating parameters (such as inlet velocity, operating pressure and temperature); (3) Automated execution scheduling for full-process end-to-end automation for efficient CFD simulations, including automatic geometry generation, boundary condition configuration, meshing, and solver invocation. Meanwhile, we summarize and analyze key weaknesses, such as insufficient physical credibility and engineering applicability limitations. Finally, we suggest future directions to possibly address these challenges, including reducing the difficulty of obtaining domain data, continuously updating domain models, and improving model interpretability. The complementary relationship between human intelligence and AI (HI-AI) will promote CFD to better serve academic and engineering applications.
- 19 March 2026
- Published by Scilight Press Pty Ltd in Smart Chemical Engineering
- Vol. 2 (1) , 3
Open AccessLarge Language Models for Automating Computational Fluid Dynamics (CFD): From Predictive Modeling and Optimization to Execution Scheduling
Artificial intelligence (AI), especially large language models (LLMs), are triggering an intelligent transformation in the field of computational fluid dynamics (CFD). LLMs can assist and even partially replace humans in executing complex fluid simulation tasks. Due to such opportunities, we systematically review recent progress and limitations of applying LLMs to the CFD. Specifically, three key applications of LLMs in CFD are surveyed, explored, and analyzed, including: (1) Predictive modeling for prediction of fluid behavior (such as real-time generation of flow field distributions) and automatic discovery of turbulence models (such as improvement of the k-ε model based on direct numerical simulation data, achieving implicit encoding of physical laws); (2) Optimization applications, including CFD model optimization (i.e., optimization of numerical model parameters and their sub-models, such as turbulence model parameters), hyperparameter optimization of machine learning models (such as learning rate, optimizer, number of neurons, and hidden layers involved in neural networks training), geometric structure optimization (such as topological optimization of fluid process equipment structures), and process operating parameters (such as inlet velocity, operating pressure and temperature); (3) Automated execution scheduling for full-process end-to-end automation for efficient CFD simulations, including automatic geometry generation, boundary condition configuration, meshing, and solver invocation. Meanwhile, we summarize and analyze key weaknesses, such as insufficient physical credibility and engineering applicability limitations. Finally, we suggest future directions to possibly address these challenges, including reducing the difficulty of obtaining domain data, continuously updating domain models, and improving model interpretability. The complementary relationship between human intelligence and AI (HI-AI) will promote CFD to better serve academic and engineering applications.