Engineering project case study
Battery SOC and SOH Estimation with Deep Learning
Research applying SABRE and RELIANT-based deep-learning methods to lithium-ion battery state-of-charge and state-of-health estimation.
Research objective
Reliable battery management depends on estimating state of charge and state of health from signals that are noisy, nonlinear and affected by operating conditions. This work explores deep-learning approaches for improving lithium-ion battery-state estimation.
Approach
Signal preparation
Prepared battery measurements for model development and consistent experimental comparison.
Model investigation
Explored SABRE and RELIANT-based deep-learning architectures for SOC and SOH prediction.
Evaluation
Compared prediction behaviour through quantitative metrics and visual analysis of model outputs.
Research communication
Documented the method and results in work titled “Enhancing SOC and SOH Estimation in Lithium-Ion Batteries through SABRE and RELIANT-Based Approach.”
Engineering relevance
Better battery-state estimates support safer operation, improved energy management and more dependable decisions in electric and autonomous systems.