Repository logo
 

Smart Management of Electric Vehicles Charging in Distribution Networks

Files

Thesis(6.72 MB)

Date

2020

Authors

Supervisor

Lie, Tek Tjing
Zamora, Ramon
Foo, Gilbert

Item type

Thesis

Degree name

Doctor of Philosophy

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

For more than a decade, global transportation and power industries have played a revolutionary role in considering alternative and sustainable solutions for internal combustion engine vehicles (ICEVs) to reduce oil dependency and environmental impact. Electric Vehicles (EVs), driven by the battery, offer unique advantages regarding emission reduction, reduced petroleum use and energy efficiency. Thus, it is noteworthy to consider the positive impacts EVs may have on power systems as numbers keep increasing in future market shares. The proliferation of EVs requires the deployment of charging facilities, scheduling strategies, and advanced power control schemes in order to manage incremental charging loads better. However, the practical and efficient application of such EV-related equipment and technologies involves challenges beyond merely upgrading the existing power grid. In the distribution network, the barriers to widespread EV adoption are (1) lack of sufficient information about EV charging profiles, (2) lack of effective control and scheduling techniques to manage EV charging loads and (3) lack of market mechanisms to maximise economic benefits. Uncontrolled EV charging can cause extra peak loading, inefficient network operation and redundant economic costs. This thesis focuses on the development of advanced modelling, scheduling, and controlling techniques that could be used within distribution networks to manage EV charging smartly. Modelling EV charging demands deals with stochastic problems related to the charging behaviours of EV users. Monte Carlo Simulation (MCS) is carried out (in manuscript 1) to demonstrate inhomogeneous charging characteristics based on a systematic investigation of the current composition of the EV fleet in New Zealand (NZ). A genetic algorithm (GA) has been applied to a smart charging strategy to mitigate the adverse impacts brought by large-scale EV integration. In a competitive market environment, EV users, utilities and charging service providers should form a community of common interests to promote widespread usage of EVs. In manuscript 2, an online scheduling strategy is proposed to investigate the significance of economic integration under the energy trade market, where all participants’ interests can be satisfied. With increasing penetrations of distributed Photovoltaic (PV) generation in addition to EVs, the intermittent and stochastic power characteristic may detrimentally affect the security of the power supply. Thus, a Dynamic Power Balance System (DPBS) with a novel control scheme is proposed (in manuscript 3) to manage dynamic power generation and consumption in the distribution network. It could be used as a supplemental measure to obtain a fast load balance response without restraining EV users and considerably curtail the risk of overloading power distribution equipment. The findings of this study revealed that current EV adoption in NZ (and many other countries in the world) is still in its early stages while the majority of the existing distribution network is not intelligent enough to integrate large-scale EVs. It was further verified that the proposed methods in this research are theoretically flexible and capable of being applied to the power grid to smartly manage EVs’ charging demand.

Description

Keywords

Electric Vehicles, Smart Charging Strategy, Control, Optimisation, Distribution Networks

Source

DOI

Publisher's version

Rights statement

Permanent link

https://hdl.handle.net/10292/13724

Collections

Doctoral Theses