FWO Senior Postdoctoral Fellow · IEEE Senior Member
Feng Guo
Physics-Informed
Battery Intelligence
Developing physics-informed models and estimation methods to make battery management systems more accurate, robust, and interpretable.
VITO & UHasselt · Belgium
Visiting Researcher · WMG, University of Warwick
I am Feng Guo, an FWO Senior Postdoctoral Fellow and IEEE Senior Member, working jointly with VITO and UHasselt in Belgium. My research connects electrochemical modelling and parameter identification with robust state estimation and fault-tolerant battery management.
Research question How can we estimate battery states accurately, reliably, and efficiently when sensing is limited, model parameters are uncertain, and sensors can fail?
Research Interests
Control-oriented electrochemical battery modelling
SPM/P2D model reduction, grouped electrochemical parameters, thermal coupling, and fast parameter identification with BOLT.
Battery state estimation and parameter identification
Robust SOC estimation, physics-guided Kalman filtering, observer design, and joint state-parameter estimation under uncertainty.
Battery safety, fault diagnosis, and fault-tolerant control
Sensor-fault diagnosis and fault-tolerant SOC estimation for reliable battery operation, extending towards interpretable ageing diagnostics and safety- and lifetime-aware control.
Physics-guided AI for battery management systems
Hybrid physics-learning models that preserve interpretability while improving robustness under real operating conditions.
Open Research
Open-source Software & Datasets
CPG-SPMT
Control-oriented, parameter-grouped electrochemical modelling with thermal effects.
GitHub
Battery Model Atlas
A curated guide to public electrochemical battery models and reproducible workflows.
GitHub
E-BatP
Traceable electrode OCP curves, diffusion, and kinetics for DFN/P2D, SPM, and SPMe models.
GitHub
RoboBATT
Real-world robot battery and actuator telemetry for battery systems research.
ZenodoConnect
Collaboration
I welcome collaboration on electrochemical model reduction, battery state estimation, physics-informed machine learning, fault-tolerant control, and open-source tools for battery research.