Date of Award

2018

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering (PhD)

Administrative Home Department

Department of Electrical and Computer Engineering

Advisor 1

Sumit Paudyal

Committee Member 1

Bo Chen

Committee Member 2

Lucia Gauchia

Committee Member 3

Seyed A. Zekavat

Abstract

Electric vehicle (EV) charging/discharging can take place in any P-Q quadrants, which means EVs could provide reactive power at any state-of-charge (SOC). This dissertation shows four-quadrant operation of EVs and aggregation of EVs for support of grid operations.

First, this work develops hierarchical coordination frameworks to optimally manage active and reactive power dispatch of number of spatially distributed EVs incorporating distribution grid level constraints. This work demonstrates benefits of coordinated dispatch of active and reactive power from EVs using a 33-node distribution feeder with large number of EVs (more than 5,000). Case studies demonstrate that, in constrained distribution grids, coordinated charging reduces the average cost of EV charging if the charging takes place at non-unity power factor mode compared to unity power factor. Similarly, the results also demonstrate that distribution grids can accommodate charging of increased number of EVs if EV charging takes place at non-unity power factor mode compared to the unity power factor.

Next, this work utilizes detailed EV battery model that could be leveraged for its four-quadrant operations. Then, the developed work coordinates the operations of EVs and distribution feeder to support voltage profile on the grid in real time. The grid level problem is devised as a distribution optimal power flow model to compute voltage regulation signal to dispatch active/reactive power set points of individual EVs. The efficacy of the developed models are demonstrated by using a LV secondary feeder, where EVs' operating in all four quadrants are shown to compensate the feeder voltage fluctuations caused by daily time varying residential loads, while honoring other operational constraints of the feeder.

Furthermore, a novel grid application, called virtual power plant (VPP), is developed. Traditional nonlinear power flow problems are nonconvex, hence, time consuming to solve. In order to be used in real time simulation in VPP, an efficient linearized optimal power flow model is developed. This linearization method is used to solve a 534-bus power system with 3 VPPs in real-time. This work also implements VPP scheduling in real-time using OPAL-RT's simulator in hardware-in-the-loop (HIL), where the loads are emulated using micro-controller devices.

Share

COinS