• Login
• Admin

Thambippillai, Manoharan Aaruththiran (2024) Electric vehicle battery pack energy management design for virtual platform based scenario testing. PhD thesis, University of Nottingham Malaysia.

PDF (Thesis - as examined) - Repository staff only until 23 April 2026. Subsequently available to Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Available under Licence Creative Commons Attribution. Download (10MB)

Abstract

Electric vehicles (EVs) are currently explored extensively by automotive manufacturers to reduce global carbon emissions. However, consumer adoption is hindered by concerns over high costs, mileage, and insufficient infrastructures in many countries. EV battery pack state monitoring methods, in particular Artificial Neural Network (ANN) based State of Charge (SOC) estimators is widely explored. Despite achieving high estimation accuracy, current designs are computationally demanding. Moreover, there is also lack of research on effective battery pack SOC estimation techniques. This requires further investigation, since EVs use large battery packs to meet their mileage and voltage needs. For testing the EV components, the sole use of physical prototypes is expensive and might be time consuming to solve the issues identified.

Thus, this research investigates these problems and aims to find solutions that can increase EV usage. Parallel-cell connected battery packs for EV high voltage power supply is introduced to enhance mileage and reduce costs. A bidirectional DC-DC converter with wide operational range and high response rate of 1.708 seconds under dynamic load conditions is integrated with this battery pack to satisfy the voltage needs. This makes the battery cells to be used purely for energy storage. An effective cell balancing topology with simple control strategy is also designed, with consistent equalisation performance regardless of cell count. The proposed cell balancing topology also performs as desired under dynamic load conditions (battery cell SOC values converge with standard deviation of 1.99 within 150 seconds). A parallel ANN based SOC estimator is designed for lithium-ion battery (LiB) cells, which can achieve similar estimation accuracy (1.82% root mean squared error) as the parallel ANN design in the literature, with reduced ANN architectural complexity (O(1.82×1012)) and small training dataset. When combined with an improved battery pack SOC estimation technique, the cells near extreme operating conditions are well represented, as compared with current techniques in the literature.

A high fidelity virtual driving platform, namely IPG CarMaker, was used to intricately model the EV and various Malaysian driving conditions, to test the performance of the proposed designs, which is economical and effective. The proposed designs reflect well on the driving conditions tested and has 34.5% more remaining usable SOC than conventional series-parallel cell connected battery pack, for the specifications used. It is also found that the proposed battery pack configuration with DC-DC converter can contribute to increased mileage and reduced cell count. Therefore, employing the proposed designs on an EV would promote effective battery pack management, increasing battery cell lifetime and reduce maintenance costs.

Item Type:	Thesis (University of Nottingham only) (PhD)
Supervisors:	Aparow, Vimal Rau Begam, Mumtaj
Keywords:	electric vehicles (EVs), carbon emissions reduction, consumer adoption, high costs, mileage concerns, infrastructure insufficiency, battery pack state monitoring
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculties/Schools:	University of Nottingham, Malaysia > Faculty of Science and Engineering — Engineering > Department of Electrical and Electronic Engineering
Item ID:	77940
Depositing User:	Thambippillai, Manoharan
Date Deposited:	27 Jul 2024 04:40
Last Modified:	27 Jul 2024 04:40
URI:	https://eprints.nottingham.ac.uk/id/eprint/77940

Actions (Archive Staff Only)

Edit View