Research

Intelligent Battery Management

Batteries are playing an increasing role in modern society, from electronics to electric vehicles and large-scale energy storage. However, major bottlenecks still limit the wider usage of batteries, particularly for safety-critical and time-sensitive applications. Our research along this direction aims to tackle these challenges by developing smart battery management systems, with focus on Lithium-ion batteries, with extension to new battery chemistries. Specific topics include how to: (1) enable the fast charging of batteries to reduce charging time; (2) develop new battery models to describe internal electrochemical-thermal-mechanical behaviors; (3) perform the diagnosis of battery health with data-driven and hybrid approaches; and (4) extend the methods to battery modules and packs while considering cell-to-cell variations. Exemplary publications:

M. Chowdhury, S.S.A. Al-Wahaibi, and Q. Lu. "Adaptive safe reinforcement learning-enabled optimization of battery fast-charging protocols." AIChE Journal 71 (1), e18605, 2025.
G. Modekwe, Q. Lu. "Physics-informed neural network with transfer learning for state estimation in Lithium-ion batteries using single particle model with electrolyte." FOCAPO-CPC, 2027.
G. Modekwe, Q. Lu. "Domain-adaptive capacity prediction for Lithium-ion batteries with pseudo-inverse gram matrix." Journal of Energy Storage, 128, 117037, 2025.

Process Control, Optimization, and Decision-Making

Process control and optimization are critical for high-performance operation of industrial process with optimality, robustness, and safety. Our research along this line focuses on developing novel control and optimization theories, methods, and algorithms to tackle the increasing complexity in dynamics, dimensions, and scale of modern processes, with applications to chemical reactions, energy systems, and distributed systems. The specific topics that we are interested into include how to develop: (1) advanced control strategies, e.g., model-predictive control and reinforcement learning (RL), to enhance operation performance, economics, adaptivity, and fault robustness; (2) optimization methods such as data-driven optimization, Bayesian optimization for enhanced computation efficiency; and (3) effective control parameter tuning/optimization with stability and safety guarantee. Examples include:

M. Chowdhury, S.A. Al-Wahaibi, Q. Lu. "Entropy-maximizing TD3-based reinforcement learning for adaptive PID control of dynamical systems." Computers & Chemical Engineering 178, 108393, 2023.
B. Jiang, W. Wang, Z. Ma, Q. Lu. "Fast charging of lithium-ion batteries using deep Bayesian optimization with recurrent neural network." IEEE Transactions on Automation Science and Engineering 22, 13150-13160, 2023.
Q. Lu, L. González, R. Kumar, V. Zavala. "Bayesian optimization with reference models: A case study in MPC for HVAC central plants." Computers & Chemical Engineering 154, 107491, 2021.

AI for Chemical and Biomedical Engineering

Machine learning has emerged as a major tool to address challenges in various chemical and biomedical applications. However, unique characterizations of these applications create unseen problems that traditional machine learning or AI cannot well address. In this thrust, we aim to develop innovative AI methods, motivated by the new challenges from these systems, to enable improved understanding and decision-making for engineering and biomedical processes. Specific topics include how to: (1) tame the high-dimensional spatiotemporal correlations for the monitoring of industrial process; (2) tackle the extreme data scarcity existing across both engineering (e.g., rare-event scenarios like thermal runaways) and biomedical (e.g., cancer cells of certainty subtypes) domains; and (3) integrate both knowledge with data insights for informed decision-making. Exemplary publications include:

B. Aryal, G. Modekwe, and Q. Lu. "Multi-level temporal graph networks with local-global fusion for industrial fault diagnosis." Measurement Science and Technology 37 (23), 235109, 2026.
J. Alam, L. Tran, D. Agrawal, A. Bari, R. Phariss, K. Gonzalez, G. Modekwe, L. Tijani, Q. Lu, J. Gomez-Pastora. "Automated detection of sickled erythrocytes in peripheral blood smear images using YOLO-based object detection." 2026.
S. Al-Wahaibi, S. Abiola, M. Chowdhury, Q. Lu. "Improving convolutional neural networks for fault diagnosis in chemical processes by incorporating global correlations." Computers & Chemical Engineering 176, 108289, 2023.

Page updated

Google Sites

Report abuse