**1. Introduction**

Charging stations or other types of energy storages need organization when they are used for special targets. Considering the grid's situation, an organized program should cover the supply and demand needs of the grid. Supply needs, such as low loss operation without side effects, and demand needs, such as abundance in the charge station, low cost, and increased power quality, are required to be fulfilled. All of these needs and similar problems, issues or remodeling methods have created a basis to providing organization for electric vehicles (EV) charge stations. Factors, such as societal, populace, grid infrastructure, and renewable conditions [1], critical power range, and financial analysis of a location, play important roles in determining the grid condition. However, supply/demand power and coordination, and the estimation of equipment is the telling factor, which needs focus to meet the incessant load demand.

The estimation of power, especially on the demand side of the grid, is very difficult [2]. Increasing or decreasing the rate of grid power does not go on determined patents, such as battery open circuit voltage estimation by Extended Kalman Filter (EKF) or similar estimation operations [3]. Supply/demand and coordinating for EVs charge station estimations have a lot in common, so they can contribute to an increase in power quality in the grid, directly when the grid infrastructure is ready. The benefits of this work are not limited to some factors, such as the decrease in green-house gas (GHG) emissions. Fuel transmission between oil refineries to gas stations consume electric power, especially in gas stations where compressed natural gas (CNG) stations use electric convertors. At present, too many researchers are trying to create and optimize algorithms to harvest fine results, but the methods and algorithms should be fed by trustworthy numerical data. In addition, the experiments are to be simulated using simulation software, such as MATLAB. In one Finnish study, to meet Finland's target of 100% renewable energy usage by 2050's (funded and presented by Ministry of Economy and Employment), the research on Energy PLAN (created by Aalborg University) modelling has focused on energy potential and generation from a perspective of economy, the rate of fuel usage, and CO2 emission according to decrease in GHG emission policy [4]. The result of this investigation shows that the energy future, based on 100% renewable energy, is entirely possible. In another study focused on Delhi, India's energy sector, the Japanese Society for the Promotion of Science (JSPS) investigated the future of energy and its optimization methods in Delhi. Their study found that integrated-systematic modelling should be used for energy flow in urban energy systems. They have used an optimal energy scenario and found an energy gap between supply and demand. This research paid attention to governmen<sup>t</sup> energy policy in improving the demand side of their energy sector [5]. However, governmen<sup>t</sup> policy and social behavior strongly influence the GHG emission in the transportation sector, in many countries, such as Austria [5]. The emission of CO2 is one of the primary problems in China and the CO2 condition will be in 2030 with the use of control or uncontrolled EVs, V2G and similar transportation systems. Studies have shown that the application of these transportation systems provide advantages, such as decreasing peak load, which is used for EV charging; generating power from renewable energy sources (RES); and reducing energy generation and EV charging costs [6]. Some of the literature has focused on the distance of charge stations with respect to the most capable grids. A Taiwanese study looked into helping customers receive better service from the grid by creating charge stations at reasonable distances in Taiwan [7]. Therefore, it is possible to determine probable locations for EV charging stations, based on the grid and transportation technology of the time, rather than based on any specific algorithm.

Several algorithm-based research are still used in smart grid and V2G operations for special conditions or purposes which are explained below. One study focused on uncertain situations with respect to V2G operation pricing, using the robust game-theory to treat uncertain issues. This algorithm was named ENTRUST (Energy Trading under Uncertainty in Smart Grid Systems) [8]. This study achieved reliability and cost e fficiency in terms of energy management. Another grid lead system is known as multi-agent energy management. In this system, all elements of demand side of the grid (home, building, industries, and vehicles) are connected. It can improve autonomy, connectivity, diversity, and appearance of the systems in the grid. Another study in Great Britain (GB) [9], created a dynamic virtual energy storage systems (VESS) to solve the dynamic frequency response. This model uses V2G and similar technology storages to notify grid whenever it needs emergency energy or whenever the rhythm of the energy flow has been disturbed. This model considered the response capacity of an EVs cluster. Researchers have furthermore developed some algorithms for EV charge stations, including the flow refueling location model (FRLM) [10] maximum covering location problem (MCLP) [11], nonlinear auto-regressive (NAR) [12], Genetic Algorithm (GA) [13], Genetic Algorithm-particle swarm optimization (GA-PSO) [14], and Genetic Algorithm-binary particle swarm optimization (GA-BPSO) [15].

Each research has coordinated points by special parameters, which is the dominant factor in supporting the system objectives. Previously described models, scenarios, and algorithms are the key to managing the grid and to increasing responsibility for EVs, charge stations and interconnected storage. The coordination of EV charging stations at first sight needs these methods to solve some pre-requisites. This paper considers Ankara's metropolitan electric infrastructure and the conditions dependent on Ankara's electricity infrastructure to determine potential locations for the charging stations. In Ankara's case, the grid has not been updated to be smart enough to include online sensors, which may be used in determined algorithms. This study attempts to determine the location by specifications of the grid infrastructure, categorizing the capacity of transformers and their other parameters. A method, based on the Genetic Algorithm is designed to verify the probability of a location meeting the requirement is shown.

The rest of the paper is organized as follows. Section 2 discusses the case study, particularly considering four locations in Ankara. Section 3 takes grids in Ankara in consideration to nominate them as the station location. Section 4 brings transformers into inspection by delineating several parameters to select transformers capable of matching the EV application. Section 5 presents the corresponding transformer indices and parameters to aid the selection process. Considering this framework, the layout of the estimation algorithm with respect to the Genetic Algorithm is explained in Section 6. Finally, conclusions are drawn in Section 7.
