Computational investigation of an adaptive PID-controlled PCM-enhanced hybrid thermoelectric-air cooling system for lithium-ion battery packs in electric vehicles

Abstract

Effective thermal management of lithium-ion battery packs remains a critical challenge for ensuring the safety, performance, and longevity of Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs). This study presents a computational investigation of a Phase Change Material (PCM)-enhanced hybrid thermoelectric-air cooling system integrated with an adaptive Proportional–Integral–Derivative (PID) controller for battery thermal management. A calibrated lumped-parameter (nodal energy-balance) thermal model was developed for a 13-series 8-parallel (13S8P) battery pack comprising 104 Panasonic NCR18650BD cylindrical lithium-ion cells arranged in a zig-zag configuration. The governing energy balance was discretised over seven thermal nodes and integrated using an explicit Euler scheme with a 1 s timestep, implemented in Python 3.11 with NumPy, SciPy and Matplotlib; model calibration used the Nelder–Mead simplex algorithm. No computational fluid dynamics solver or spatial mesh was employed. The model was validated against published experimental data, achieving a coefficient of determination ( R 2 ) of 0.9515 and a mean absolute percentage error of 6.2% (7.3% excluding the trivially zero initial condition). Five distinct cooling modes were systematically evaluated: natural convection, forced convection, thermoelectric convection, hybrid (forced + thermoelectric), and a novel PCM-enhanced hybrid incorporating RT42 paraffin wax. Comprehensive analyses were conducted across discharge rates (5–30A), airflow speeds (2–15 m/s), four battery pack configurations, five aging states, four driving cycles, and parametric sensitivity studies. The results demonstrate that the PCM-enhanced hybrid mode achieves a peak temperature of 48.9 °C at 30A discharge, representing a 49.8% reduction compared to the standard hybrid mode (97.5 °C), while simultaneously reducing temperature non-uniformity by 58.6%. The adaptive PID controller, which dynamically switches among all five cooling modes based on real-time thermal conditions, achieves 29.75 ± 0.09 % energy savings compared to a threshold-triggered baseline (mean ± standard deviation over 20 seeded repetitions of the Urban Dynamometer Driving Schedule, UDDS) while maintaining the temperature below the 60 °C safety limit. A parametric sensitivity analysis identifies PCM mass as the most influential design parameter (−14.1% sensitivity index). Battery aging to 2000 cycles increases the peak temperature from 42.4 °C to 69.0 °C, exceeding the safety threshold and necessitating age-aware control recalibration. These findings provide comprehensive design guidelines for energy-efficient and thermally robust battery thermal management systems for EV applications.