Battery discharge stage system improves remaining use time prediction
Relative Discharge Stage (RDS) Classification: A Practical Indicator of Battery Discharge Progress
Machine Learning
Summary
Predicting exactly how much time a battery will last before running out is tricky because future use is unpredictable and changes a lot. The authors created a new way to classify the battery’s remaining discharge into five easy-to-understand stages instead of trying to predict exact minutes. This approach doesn’t need to know future battery usage, only current measurements and estimates. They combined physics-based battery models with a simple neural network, and tests showed over 80% accuracy on real battery data under different conditions.
What this means in practice
- •For battery engineers: Use the relative discharge stage to track battery discharge progress without needing future load predictions during battery management.
- •For electric vehicle maintenance teams: Monitor battery health and predict recharge needs under varying driving and temperature conditions more reliably using relative discharge stages.
Authors
Khoa Tran, Tri Le, Hung-Cuong Trinh, Hung Tran-Nam
Abstract
Accurate remaining discharge time (RDT) prediction is challenging in real-world battery applications because future load profiles are unknown and highly dynamic. To address the uncertainty of continuous RDT regression, this paper introduces Relative Discharge Stage (RDS), a battery-management indicator that represents the remaining discharge condition using five interpretable classes: Normal, Good, Moderate, Low, and Recharge Required. Unlike state of charge (SOC), which reflects the current charge level, RDS characterizes the remaining discharge process without requiring future-current information during inference. A physics-informed RDS classification framework is proposed, combining SOC estimation with lightweight temporal learning. The SOC-estimation component includes second-order ECM state and terminal-voltage prediction, hysteresis and OCV temperature correction, core-temperature estimation, and AEKF state correction, supported by OCV evaluation, online STC-ECM parameter adaptation, and pretrained neural residual-voltage correction. The measured current, terminal voltage, surface temperature, and estimated SOC are arranged into a sliding observation window and processed by a lightweight temporal convolutional network. Experiments on two public lithium-ion battery datasets demonstrate robust RDS classification, with accuracy exceeding 80% under varying load and thermal conditions.