Simulation study on optimal placement and sizing of Battery Switching Station units using Artificial Bee Colony algorithm J.J. Jamian a,⇑ , M.W. Mustafa a , H. Mokhlis b , M.A. Baharudin a a Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, Malaysia b Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia article info Article history: Received 30 July 2012 Received in revised form 10 September 2013 Accepted 12 October 2013 Keywords: Artificial Bee Colony Battery Switching Station Distribution generators Optimal placement Power loss improvement abstract The improvements on DC electrical motor and battery technologies have increased the industries’ and users’ interest in Electrical Vehicle (EV). In line with this, multiple types of Charging Stations (CS) have also been introduced to support the EV’s growth in the vehicle market. Three types of CS units that are available nowadays are Levels 1, 2 and 3 (DC Charging type). However, there are some drawbacks on these CS types either in terms of charging time (for Levels 1 and 2) or the impact to the system perfor- mance (Level 3). Thus, the Battery Switching Station (BSS) is used to tackle all these problems. In the BSS, the customers’ current batteries are swapped with fully charged batteries by the automated system and the charging process of those batteries will be done at a later time by the BSS. Therefore, the long charg- ing time issue can be evaded. However, inappropriate placement of BSS in the distribution network will give enormous impact to the distribution network due to the lower voltage level in distribution system. Two methods have been used to place the BSS in this study, which are, analytical and meta-heuristic (Artificial Bee Colony (ABC)) methods. The ABC is chosen due to the simplicity of the algorithm in finding the optimal location of BSS. For further improvement, the ABC is employed to find the optimal size of BSS and DG output. The results show that the analytical and optimization methods gave similar results in finding the optimal location of BSS, which is better than randomization of BSS placement. Furthermore, the analysis on optimal size of BSS and DG output (simultaneously) gives the lowest power loss compared to other analyses. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction The current advancement on DC motors and battery technolo- gies has given positive effects on the growth of Electrical Vehicle (EV) industry. Most of the well known automobile companies such as Jaguar, Volkswagen, BMW, Nissan, Toyota, Honda, Proton and others [1–3] have started to produce their own EV model especially the Plug in Hybrid Electric Vehicle (PHEV) model and electric- based vehicle (PEV). With the growing number of EV deployment in the future, some critical problems that occurred during the fuel base-vehicle (FBV) era, such as the fuel price and environmental is- sues, can be solved [4–6]. However, the problem of ‘‘charging time duration’’ arises when people migrate from FBV to EV. For short distance usage, the EV can be charged at home during the night time with cheaper price (for countries with dynamic electricity tar- iffs). For example, in the United Kingdom, one of the PEV model, Citroen ev’ie, required only 6 h to fully charge the battery using a 13A socket with the charging cost of less than 0.95p [7]. The car can travel up to 75 miles and it is feasible for short distance daily users. On the other hand, for long distance usage, 6 h is a long time for the user to wait at the charging point before continuing their journey. Therefore, a faster and a more efficient charging station are required. The Society of Automotive Engineers (SAE) J1772 has catego- rized the Charging Station (CS) into three categories, which are, CS Levels 1, 2 and 3 [8–10]. Levels 1 and 2 use the AC/DC charging approach to charge the battery with the voltage level of 120 and 208/240 V AC , respectively, while for Level 3, the high voltage and fast-rate DC Charging Slot (DCCS) approach is used to charge the EV (also known as DC Faster Charger). The main advantage of using DCCS approach is the short charging time, which is similar or faster than the conventional gasoline station. Aggeler et al. [11] have dis- cussed the main requirements needed by the Power Electronic (PE) architecture to achieve Ultra-Fast DC Charge for EV such as the DC output of PE must be in the range of 100–600 V, charging time less than 10 min and maximum current and voltage ripple are 1% and 5% respectively. From the research, all the requirements of DCCS can be fulfilled either using Low Frequency Conversion Architec- ture with non-isolated three phase buck converter or High 0142-0615/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijepes.2013.10.009 ⇑ Corresponding author. Tel.: +60 13 3854888; fax: +60 7 5566272 (M.W. Mustafa). E-mail addresses: jasrul@fke.utm.my (J.J. Jamian), wazir@fke.utm.my (M.W. Mustafa), hazli@um.edu.my (H. Mokhlis), mariff@fke.utm.my (M.A. Baharudin). Electrical Power and Energy Systems 55 (2014) 592–601 Contents lists available at ScienceDirect Electrical Power and Energy Systems journal homepage: www.elsevier.com/locate/ijepes