Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Electrical and Computer Engineering

Committee Chair

Naber, John F., 1961-

Committee Co-Chair (if applicable)

McIntyre, Michael L.

Committee Member

Graham, James

Committee Member

Bai, Lihui


Due to the recent inflow of Electric Vehicles (EVs) to the automobile market, new concerns have risen with respect to the additional electrical load and the resultant effects on an overloaded electric grid. Either for convenience purposes or possibly necessity due to limited electric range on EVs, some EV owners may desire to charge their EV while at work in addition to charging at home. These forward-thinking daytime charging providers are typically Commercial and Industrial (C&I) electric ratepayers, or other large electric consumers which constitute the majority of businesses, shopping centers, academic campuses and manufacturing facilities. Increased electricity consumption due to EV charging activity results in higher electricity costs due to differences in the billing structures between residential and C&I electric ratepayers. Therefore, it is beneficial to the EVSE charging provider to minimize charging activity around peak demand periods which would result in lower electrical costs overall. A solution is developed that can provide this control without creating a nuisance to electric vehicle owners since EV charging demand is somewhat inelastic due to range anxiety. The primary objective of the research detailed in this dissertation is to develop a novel demand side management system for monitoring the peak demand of commercial time-of-day electric ratepayers that cost effectively predicts and controls electric vehicle charging during peak demand periods. This objective is achieved, therefore confirming the hypothesis that such a system can provide cost and demand benefits to forward-thinking commercial electric ratepayers that provide daytime charging capabilities. This work proposes and evaluates a novel Power Monitoring and Control System (PMCS) that can be implemented at C&I EV charging locations to minimize or eliminate the negative impacts of charging electric vehicles at the workplace in C&I environments. Operation of the PMCS begins by forecasting electrical demand in advance of every 15 minute demand interval throughout the day. The forecast is generated using an artificial neural network and a number of input data streams. Electrical demand has been shown to correlate well with weather data such as temperature and dew point. Therefore, using those measurements along with a date and time stamp, and historical electrical demand measurements, a highly accurate forecast for the following 15-minute demand interval was achieved. From that forecast, the number of EV charging stations that may be active, without the chance of creating new electrical demand peaks, is calculated. Finally, the forecast is then used to properly schedule EV charging activity so that electrical demand peaks can be avoided but charging activity is maximized. The avoidance of charging activity at or near peaks in electrical demand results in lower total electric costs associated with the charging process. The final design was implemented in an EV charging testbed at the University of Louisville and data was collected to verify the operation and performance of the PMCS. With a properly designed scheduling and prioritization control algorithm, increases in electrical demand and associated costs are limited to the error in the forecasting algorithm used for predicting electrical demand levels. The final design of the forecasting algorithm results in a mean absolute percent error of 0.02% to 0.08% in the electrical demand forecast. This corresponds to approximately 3 to 10 kVA of error in electrical demand. Taking this error into account, total cost of charging several EVs is reduced by nearly 90%. Furthermore, for scenarios where there are several more electric vehicles requiring charge than there are charging stations available, several scheduling algorithms are presented in an attempt to minimize the total processing time required for completing all charging transactions.