Numerical and artificial neural network inspired study on step-like-plenum battery thermal management system

Numerical and artificial neural network inspired study on step-like-plenum battery thermal management system

This study leverage numerical simulation (NS) and artificial neural network (ANN) capabilities to carry out additional investigations on step-like plenum battery thermal management system (BTMS). Different cooling strategies have been developed over the years in BTMSs’ design. Yet, air-cooling strat...

Full description

Saved in:
Bibliographic Details
Main Authors: Olanrewaju M. Oyewola, Emmanuel T. Idowu
Format: Article
Language:English
Published: Elsevier 2024-11-01
Series:International Journal of Thermofluids
Subjects:
CFD
Machine learning
ANN
BTMS
Step-like
Plenum
Online Access:http://www.sciencedirect.com/science/article/pii/S2666202724003379
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study leverage numerical simulation (NS) and artificial neural network (ANN) capabilities to carry out additional investigations on step-like plenum battery thermal management system (BTMS). Different cooling strategies have been developed over the years in BTMSs’ design. Yet, air-cooling strategies still remains relevant, especially in battery-powered aircrafts, where light-weight is important and air is the preferred cooling fluid. Hence, additional study become necessary especially on the step-like plenum design to provide more insight on the performance of the design by considering several number of step, varied air inlet temperature and velocity. Computational fluid dynamics (CFD) approach was employed to obtained results for different number of step; Ns = 1,  3,  4,  7,  9,  15 and 19, varied air inlet temperature; Ti = 278,  298 and 318 K, and varied air inlet velocity; Vi = 3,  3.5,  4,  5 and 6 m/s. Artificial Neural Network (ANN) approach was then employed to predict the BTMSs’ performance for additional values of Ti and Vi. Minimum temperature (Tmin), maximum temperature (Tmax), maximum temperature difference (ΔTmax) and pressure drop (ΔP) were computed. By comparing the CFD results with the result predicted by the ANN, the percentage difference, for the entire dataset were 0.01 %, 0.005 %, 1 % and 0.14 % for Tmax, Tmin ΔTmax and ΔP, respectively. Based on the optimum design parameters predicted using ANN, for Tmax = 299.24 comprises Ns = 4, Vi = 6 m/s and Ti = 278 K, while for ΔP, comprises Ns = 1, Vi = 3 m/s and Ti = 318 K.
ISSN:2666-2027

Similar Items