Reducing Probability of Transformer Failure by Managing EV Charging in Residential Parking Lots Milad Soleimani * , Mohammad Khoshjahan, Mladen Kezunovic Department of Electrical and Computer Engineering Texas A&M University College Station, TX, United States. * msoleimani@ieee.org Abstract—The power of electric vehicle (EV) chargers is considerable and high penetration of EVs may lead to overloading and thermal stress for utility transformers. Large buildings usually are connected to the grid through a transformer. By managing EV charging in the building parking lots, the probability of transformer failure may be reduced. We propose a controller to manage the charging of the EVs to reduce the probability of transformer failure without the involvement of distribution grid operator. In order to test the proposed framework, a use case is developed using real and synthesized data from College Station, TX, United States. Index Terms—Electric vehicle, electric transformer, loss of life, hazard of failure, fuzzy control. I. INTRODUCTION Electric vehicles (EVs) provide clean and environmental- friendly features and the increased public awareness have caused considerable increase in EV deployment [1]. This increase can cause repetitive utility transformer overloading and lead to accelerated ageing and increased probability of failure. Distribution transformers are not usually equipped with elaborate overload protection and monitoring devices and this overloading stress may remain undetected until it leads to transformer failure. Apartment complexes are subject to this increased penetration of EVs. Due to cultural, financial, and social similarities in the residents of an apartment complex, the rapid increased penetration of EVs in some affluent neighborhood is quite likely. If the number of residential dwellings in an apartment complex is high enough, one transformer may be assigned for the whole building. Therefore, the charging of all EVs in the building can be managed using a single EV charging management system. The adverse impact of EV charging on the transformer life is studied in [2], [3] and it is shown how increased penetration of EVs may lead to accelerated ageing of transformers and its economic impact is quantified. The impact that different EV penetration levels can have on transformers is studied in [4] and it is illustrated that even low level penetration can be troublesome. Deploying PV generation and battery energy storage to mitigate the transformer loss of life is introduced in [5] and [6]. The methods introduced in [5], [6] require the presence of PV generation and battery energy storage in the consumer location which may not be readily available and is costly to provide. In [7] and [8], a rule-based method is proposed in order to mitigate the impact of EV charging on transformer by peak shaving. In the authors’ published work [9], a framework is proposed to manage the transformer risk of failure as well as loss of life using a charging scheduling system in the consumer location and a decision-support tool in the grid operator location. They interact in real-time to manage the charging. In the system proposed in the authors previously published work in [9], it is assumed that only one EV at a time can be charged in each building. Thus, this method cannot work in a parking space where several EVs need to be charged at the same time. Also, a communication infrastructure between the building and grid operator is needed. Structure and functionality of the controller in [9] is serving an operating condition for one residential building with one charging spot. The charging necessity factor as well as the decision-making algorithm used in [9] led to unready EVs in multiple occasions in the case study. In this paper, this index and the decision-making algorithm are revised to address the issue when multiple EVs are involved, which significantly changes the structure and functionality of the controller. The main contribution of this paper is in developing the scheduling system for a residential apartment building parking with multiple EVs. Here, we include the capability of managing the charging of multiple EVs at the consumer location when a transformer is assigned specifically to the apartment building without a need for consumer-operator communications and operator decision tool. The economic impact of implementation of the proposed solution for several levels of penetration of EVs is studied. The method is tested using a realistic use case that is developed by deploying real and synthesized data. The remainder of the paper is organized as follows. Transformer loss of life and hazard of failure are quantified in This material is based upon work supported by the Department of Energy, Office of International Affairs and Office of Electricity under Award Number DE-IA0000025. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.