دانلود رایگان ترجمه مقاله تخصیص همزمان پارکینگ وسایل نقلیه الکتریکی و منابع تجدیدپذیر توزیع شده – الزویر 2017

تخصیص هم زمان پارکینک های خودرو های الکتریکی و منابع تجدیدپذیر توزیعی در شبکه های توزیع برق هوشمند

Simultaneous allocation of electric vehicles’ parking lots and distributed renewable resources in smart power distribution networks

Abstract

Electric vehicles (EVs) and distributed renewable resources (DRRs) are introduced to achieve three of the most pivotal objectives of this century: using environmentally-friendly energy resources, reliable supply of the load demand, and sustainable development of power systems. To achieve the aforementioned goals, simultaneous utilization of DRRs and EVs should be implemented in a scheduled manner. In this paper, we propose a two-stage approach for allocation of EV parking lots and DRRs in power distribution network. Our method considers both the economical benefits of parking lot investor and the technical constraints of distribution network operator. First, the parking lot investor offers the candidate buses for installing the parking lot to the distribution network operator based on economic objectives. Then, the distribution network decision-making is obtained to reduce loss of system. The proposed framework not only improves the distribution network loss, but also ameliorates the availability of the parking lot from the economical point of view. In order to solve the formulated optimization problem, we utilize two optimization techniques. Genetic algorithm (GA) and particle swarm optimization (PSO) algorithm are used for the distribution network loss minimization purpose. Besides, we model the EV parking lot by expanding single EV probabilistic model. The performance of the proposed method is evaluated by allocating DRRs and EV parking lots simultaneously on the IEEE standard distribution test system. This system is bus 2 of Roy Billinton Test System (RBTS).

1 Introduction

1-1 Motivation

Growing awareness of energy and environment, and the demand for a reliable, secure, and sustainable power grid, lead to the evolution of smart grid (SG) as a reliable means of electricity distribution. Novel technologies are introduced by SG, such as selfhealing frameworks for smart distribution systems management (Zidan & El-Saadany, 2012), and smart buildings with vehicle-togrid (V2G) capability (Lamedica et al., 2015). In this context, electric vehicles (EVs) play a pivotal role (Brown, Pyke, & Steenhof, 2010). Brady and O’Mahony proposed a charging profile modeling for electric vehicles in (Brady & O’Mahony, 2016). Their model is based on real-world EV charging data. Additionally, the increasing load demand and the limits on energy resources involve incorporating renewable resources. Enabling the SG also revolutionizes the characteristics of distribution networks (Brown, 2008). Based on US Energy Information Administration (EIA) assessment, energy demand will increase by 56% from 2010 to 2040 which is driven by economic development (EIA, 2013). Therefore, governments have proposed distributed renewable resources (DRRs) and EVs as technological solutions for solving sustainability issues. Unscheduled and independent utilization of DRRs and EVs may bring irrefutable challenges for the smart distribution network. The future cities will require to deploy novel technologies such as electric vehicles and renewable resources to improve social welfare, reduce environmental emissions, and move toward electrified transportation networks. This also will lead to sustainable development of the power distribution networks.

1-2 Literature survey

The effects of EVs and DRRs on the future power systems are investigated in several papers (e.g., references (Amini, Nabi, & Haghifam, 2013; Clement-Nyns, Haesen, & Driesen, 2010; Kempton & Tomic, 2005; Shaaban, Atwa, & El-Saadany, 2013; Subramanian et al., 2013; Tuttle & Baldick, 2012)). In (Clement-Nyns et al., 2010), the effect of EV charging on a residential distribution grid was studied. However, the effect of simultaneous utilization of DRRs on the EV charging was not addressed. The value of using DRRs in distribution networks includes investments deferral, loss reduction, and improving the network reliability (Shaaban et al., 2013). In (Subramanian et al., 2013), the advantages of scheduled aggregation of deferrable loads and storages were investigated. In (Amini et al., 2013), a general smart load management framework was introduced based on multi-agent systems considering DRRs and responsive demands. In (Kempton & Tomic, 2005), the potential benefits of EV’s participation in different types of electricity markets such as regulation market and spinning reserves were studies. Alizadeh et al. in (Alizadeh et al., 2014) proposed a stochastic model for EV load estimation. Their model facilitates the integration of EVs in future power networks. Furthermore, energy security and the cost savings related to oil consumption reduction were considered as advantages of EV utilization (Tuttle & Baldick, 2012). In (Amini & Karabasoglu, 2016), we proposed a comprehensive framework to model the interdependent nature of power and transportation networks. This model considers the limits of transportation network as well as the power systems operational constraints by solving multi-objective route optimization and optimal power flow problem iteratively. A review of optimization techniues for electric vehicles charging infrastructure is provided in (Rahman et al., 2016). Previous studies have focused on the distributed generation (DG) allocation using various optimization algorithms including Genetic algorithm (Mohan & Ramesh, 2012), and particle swarm optimization (Bhumkittipich & Phuangpornpitak, 2013). Besides, some studies proposed methods for parking lots allocation in distribution networks. A congestion management method is developed in (López, Martín, Aguado, & de la Torre, 2013) considering high penetration of EVs. In (Moradijoz, Parsa Moghaddam, Haghifam, & alishahi, 2013), a multi-objective optimization problem is formulated and solved for optimal allocation of parking lots. In (Amini, Karabasoglu, Ilic, Boroojeni, & Iyengar, 2015), a time series forecasting approach is utilized to reduce the error of EV charging demand estimation. In (Amini, Kargarian, & Karabasoglu, 2016) a decoupled demand forecasting approach based on ARIMA model was proposed for EV charging demand. Further, a chance-constrained unit commitment problem utilizing the predicted demand was formulated to improve the accuracy of stochastic power system operation (Amini et al., 2016). In (Lam, Leung, & Chu, 2014), EV parking lot placement problem is solved in the future smart city environment (Lam et al., 2014). According to (Liu, Wen, & Ledwich, 2013), the sizing and placement of EV charging station problems have been solved separately. Furthermore, simultaneous placement of conventional and renewable distributed resources can be tackled by fuzzy multi-objective optimization (Mohsenzadeh & Haghifam, 2012). The electrical protection requirements of utilizing EV charging stations have been reviewed in (Sanchez-Sutil, Hernández, & Tobajas, 2015). In (Zakariazadeh, Jadid, & Siano, 2015), the integration of high penetration of EVs and distributed generations on distribution network is studied. In (Neyestani et al., 2014) (Amini & Islam, 2014) , EV parking lots allocation is obtained considering power distribution network’s constraints. The utilization of DRRs will lead to active distribution networks.Optimum location and charging of electric vehicles have been investigated in (Sachan & Kishor, 2015, 2016). Recent studies on the allocation of EV parking lots and DRRs can be classified into three categories: integrating EVs in a grid with DRR or conventional distributed generation penetration (El-Zonkoly & dos Santos Coelho, 2015; Fazelpour et al., 2014; Honarmand, Zakariazadeh, & Jadid, 2014), allocation of EV parking lots in the power systems (Kazemi et al., 2016; Moradijoz et al., 2013; Neyestani et al., 2014), and allocation of DRRs and DGs in power systems (Bhumkittipich & Phuangpornpitak, 2013; Mohan & Ramesh, 2012; Mohsenzadeh & Haghifam, 2012). In (El-Zonkoly & dos Santos Coelho, 2015), a multi-objective algorithm was developed to find the optimal number of parking lots, their location and size. This study used artificial bee colony (ABC) and firefly algorithms as two effective methods for optimization (El-Zonkoly & dos Santos Coelho, 2015). Arasteh et al. in (Arasteh et al., 2016) utilized SoS-based multiobjective approach to deal with distribution network planning. Honarmand et al. in (Honarmand et al., 2014), proposed an energy resources management framework considering practical constraints, DRR generation forecasting errors, and EVs owner satisfaction. Their model is applicable to microgrid and did not consider the allocation of EV parking lots (Honarmand et al., 2014). In (Kazemi et al., 2016), an allocation approach is proposed for EV parking lots considering the welfare of EV owners. The aforementioned studies considered EV parking lot allocation or DRR/DG allocation as their optimization problem. However, in this study we allocate DRRs and EV parking lots simultaneously. Our proposed method is two-tier to satisfy EV parking lot investor and distribution network operator constraints and objectives, i.e. we first consider the parking lot investor’s decision parameters. We next solve the optimization problem from the utility perspective for optimal allocation of EV parking lots and DRRs. It is worth noting thatindependent allocation of EVparking lots and DRRs may lead to sub-optimal solutions.