Multipulse HPPC for Current-Scheduled ECM Parameterization and SoC Estimation in EV Batteries
- 1Department of Electrical and Computer Engineering, Mahindra University, Hyderabad 500043, Telangana, India
- 2Centrale Lille Institute, University of Lille, 59000 Lille, France
- 3Department of Electrical and Electronic Engineering Technology, Faculty of Engineering and Built Environment, University of Johannesburg, Johannesburg 2092, South Africa
Abstract
Accurate state-of-charge (SoC) estimation is central to safe and efficient Battery Management Systems (BMS) in electric vehicles. This study evaluates a multipulse Hybrid Pulse Power Characterization (HPPC) protocol using three discharge rates (0.5 C, 1 C, and 2.5 C) and two regenerative-charge rates (−0.5 C and −0.4 C). Three discharge-derived first-order RC parameter sets are scheduled within an extended Kalman filter according to instantaneous current magnitude and tested on a 54.6 V, 20.8 Ah NMC battery pack. A matched-input dynamic record is used to compare the single-pulse and multipulse HPPC-derived estimators. Because the reference SoC is nominal-capacity Coulomb counting using the same measured current, the reported SoC errors quantify agreement with that trajectory rather than independently measured ground-truth SoC. The multipulse parameterization reduced SoC RMSE from 2.1354% to 1.4065%, MAE from 1.7001% to 1.1138%, and MaxAE from 3.8951% to 3.6014%, while R2 increased from 0.98882 to 0.99515. For voltage estimation, RMSE decreased from 0.3145 V to 0.2409 V and MAE from 0.2379 V to 0.1532 V, but MaxAE increased from 1.7800 V to 2.1770 V. The results therefore show lower aggregate estimation error on the shared dynamic record without establishing independent absolute SoC accuracy or universal ECM fidelity.
Related
- Multi-Stage State-of-Health Estimation for Lithium-Ion Batteries Based on an Adaptive SHAP-Guided Dual-Attention LSTM
- State of Charge Estimation for Lithium-Ion Batteries Under Wide Temperature Ranges Using Error Feedforward of Sparrow-Optimized Back Propagation Neural Network
- Physics-Informed Explainable Machine Learning for Hybrid State of Health and Remaining Useful Life Estimation of Lithium-Ion Batteries
- Considering Calendar-Aging High-Precision State of Health Estimation and Active Management Technology for Energy-Storage Batteries
- Handling Capacity Regeneration Effects for Accurate SOH Estimation in Lithium-Ion Batteries via Optimized Transformer
BibTeX
@article{bhavanam2026multipulse,
title = {{Multipulse HPPC for Current-Scheduled ECM Parameterization and SoC Estimation in EV Batteries}},
author = {Yaminidhar Reddy Bhavanam and Sreedhar Madichetty and Abdelkader EL KAMEL and Bonginkosi Thango},
journal = {Batteries},
year = {2026},
month = nov,
doi = {10.3390/batteries12110465},
url = {https://www.mdpi.com/2313-0105/12/11/465},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Multipulse HPPC for Current-Scheduled ECM Parameterization and SoC Estimation in EV Batteries
AU - Yaminidhar Reddy Bhavanam
AU - Sreedhar Madichetty
AU - Abdelkader EL KAMEL
AU - Bonginkosi Thango
PY - 2026
DA - 2026/11/04
JO - Batteries
DO - 10.3390/batteries12110465
UR - https://www.mdpi.com/2313-0105/12/11/465
AB - Accurate state-of-charge (SoC) estimation is central to safe and efficient Battery Management Systems (BMS) in electric vehicles. This study evaluates a multipulse Hybrid Pulse Power Characterization (HPPC) protocol using three discharge rates (0.5 C, 1 C, and 2.5 C) and two regenerative-charge rates (−0.5 C and −0.4 C). Three discharge-derived first-order RC parameter sets are scheduled within an extended Kalman filter according to instantaneous current magnitude and tested on a 54.6 V, 20.8 Ah NMC battery pack. A matched-input dynamic record is used to compare the single-pulse and multipulse HPPC-derived estimators. Because the reference SoC is nominal-capacity Coulomb counting using the same measured current, the reported SoC errors quantify agreement with that trajectory rather than independently measured ground-truth SoC. The multipulse parameterization reduced SoC RMSE from 2.1354% to 1.4065%, MAE from 1.7001% to 1.1138%, and MaxAE from 3.8951% to 3.6014%, while R2 increased from 0.98882 to 0.99515. For voltage estimation, RMSE decreased from 0.3145 V to 0.2409 V and MAE from 0.2379 V to 0.1532 V, but MaxAE increased from 1.7800 V to 2.1770 V. The results therefore show lower aggregate estimation error on the shared dynamic record without establishing independent absolute SoC accuracy or universal ECM fidelity.
SN - 2313-0105
ER -