Novel approaches to optimize and mitigate the impact of high penetration level of electric vehicles on the distribution network

Abstract

In recent years, the integration of Electric Vehicles (EVs) into the distribution network is studied intensively. One of the major concerns is that EVs consume lots of energy during a short period when most of them are simultaneously connected to the grid (during the night or during working hours). Therefore, in case the consequence of simultaneous charging is not resolved, undesired peak loads may appear on the distribution network. This is the reason why research is actually focusing on optimization and control algorithms to be executed at different levels of the network. Since the number of EVs increases drastically, the power demand during specific periods of the day would cause severe issues on the network. As a matter of fact, high peak demand may exponentially reduce the lifetime of the transformers and may damage some elements on the network. Therefore, severe voltage drop and blackouts of some regions or on the complete network can be the consequences. To solve the problem, many studies were conducted to reduce the impact of integrating EVs on the network. Their main goal was to limit the peak demand created by the EVs in order to protect the distribution grid from any damages. To do so, many optimization techniques and control strategies were used to mitigate the impact of EVs on the network. The main goal of using optimization techniques is to schedule the charging and discharging of EVs during their connection period, in which their charging will be shifted to periods where the demand on the network is low. For this purpose, Demand Response Programs (DRP) are used to incite the end-users consuming during low electricity prices and reducing their consumptions during high prices. As its name indicates, the DRP uses the supply and demand models to price the electricity depending on the power consumption of the end-users and the available power generated by the power utility. The electricity price may vary in time, in which in periods when the consumption is high, the electricity price will be high in order to let the end-users shift their power consumption to other periods when the price and the consumption are low. This strategy will help the power utility to control the power demand of the end-users and reduce the burden in certain periods in a day. Despite the advantages of using the DRP in controlling the total load demand on the distribution network, it is still limited by its long-time response (one day ahead pricing, hourahead pricing, etc.). Therefore, DRPs are useful when the speed of response to a certain unfavorable situation does not require an instant action or intervention, in other words, when losses are to be reduced, customer expenses are to be minimized, and operator benefits are to be maximized. DRPs can’t eliminate any sudden variation of the load which may produce a blackout or damage some components or even reduce their lifetime. Researchers have indeed suggested multiple methods for finding fast response solutions in order to prevent any technical or economic undesired phenomena occurring on the network. Literature review shows that these existing studies offer a solution for a partial aspect of the problem. It is seen that published results show the improvement on either technical or economic or implementation issues, but to the best of our knowledge, existing approaches cannot find a global optimum, considering all the electrical distribution aspects (pricing, maintenance, customer satisfaction, technical losses, stability, availability, etc.). The suggested approach has a wider view of the problem since it considers EV penetration at different levels of the network, it considers if the network’s infrastructure is conventional or modernized, it considers the lifetime of the distribution transformers, it minimizes losses and it introduces a collaborative algorithm for finding the global optimum solution. Comparative studies show the advantages of the suggested approach compared to the existing ones. Major findings in this thesis can be summarized as follows (i) the total load demand on the transformer respects its limit, which will increase its lifetime, (ii) the voltage profile respects the limits on the network and the transformers, (iii) the energy losses on the network are reduced, (iv) the depreciation cost of the network is reduced, (v) the revenue of the power utility and distribution system operator is increased, (vi) and finally the end-users are satisfied because the proposed strategies help them to reduce their electricity cost. Therefore, both end-users and power utility are satisfied.

Date	8 Jul 2019
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Ambrish Chandra (Supervisor) & Maarouf Saad (Co-supervisor)