An approach for estimation of optimal energy flows in battery storage devices for electric vehicles in the smart grid

. While the number of the vehicle actuated with liquid fuels are settled, the count of electric vehicles is increasing. For the present moment most of them are scheduled for daily urban usage. This paper presents an analytical approach for estimation of the impact of electrical vehicle (EV) battery charging on the distribution grid. Based on the EV charge profile, load curve and local distributed generation the grid nodes, the time variation of grid parameters is obtained. A set of typical load profiles of EV charging modes is studied and presented. A software implementation and a 24h case study of low voltage distribution network with EV charging devices is presented in order to illustrate the approach and the impacts of EV charging on the grid. In the current paper an approach using variable nonlinear algebraic equations for dynamic time domain analysis of the charge of the electric vehicles is presented. Based on the results, the challenges due to EV charging in distribution networks including renewable energy sources are discussed. This approach is widely applicable for various EV charging and distributed energy resources studies considering control algorithms, grid stability analysis, smart grid power management and other power system analysis problems.


Introduction
In order to preserve the environment through a reduction of harmful emissions, in our days the usage of the electric vehicles is growing.And thus it is established that their charge stations are a major challenge for low voltage distribution networks stability [1,2].The simultaneous operation of multiple EV charge stations represents a significant load for which most of the distribution grids are not designed for.For this reason, the influence of various EV charge profiles on microgrids is a topic of interest [3].In the case of unidirectional charging devices (the ones that only consume power and do not provide grid support) the charging process can be represented as a distributed load to reduce grid congestion.The growing number of EV leads to a considerable increase of the load in distribution networks and difficulties in managing overloads and dispatching of the grid.
A novel approach for charging stations control in order to decrease peak loads and optimize energy flows in distribution networks is presented in this paper.Objective of the proposed solution is the implementation of a schedule which supports the grid at peak load periods using the energy storage elements of electric vehicles.This can be done by using bidirectional charging devices for EV -in this way they can consume energy in periods of excess production and supply it to the local grid afterwards.As a result, balancing of the daily energy flows can be achieved.

Description of the studied microgrid
Recently a variety of techniques and methodologies for power system analysis exist.Even in a system with multiple branches, the classical analysis of energy flows can provide a precise solution to the given problem, at the cost of increased computational time.Nevertheless limitations based on the availability of the electric vehicles are imposed -their usage is based on different repetitive patterns in weekdays and the weekend.
In the current paper an approach using variable nonlinear algebraic equations for dynamic time domain analysis of the charge of the electric vehicles is presented.On Fig. 1 is shown the studied system: a threephase 0,4 kV distribution network with 21 nodes located in a small village near the city of Sofia [4].The grid is equally distributed, the loads are considered symmetrical in all phases and with power factor correction (predominantly active load) therefore the compensation of reactive power is not taken into consideration in this study.The minimal and the maximal load of the system without EV charging are respectively Pmin = 33,2 kW and Pmax = 110,8 kW.At node 10 and 18 of the grid are connected two photovoltaic (PV) arrays: one with a peak power PG1 = 30 kW connected at node 10 and the second is connected at node 18 and has a peak power PG2 = 50 kW.
The values used for analysis of the electrical system are the magnitudes and phase angles of the voltages, the active and reactive powers.The following equations describe the state of each node in the studied microgrid [3]. Where

Types of electric vehicles present in the system
For this study are considered four common models of electric vehicles: Mitsubishi i-MiEV family, Nissan Leaf, Tesla Model S and BMW i3.Those models were also chosen because of the fact that their characteristics and charging profiles (in normal charge and fast charge modes) are accessible from numerous sources including documentation provided by the producers.In future, the number EV in urban areas is expected to increase, each model having different charge profile, therefore the impact of an increased number of different models and charging profiles can be examined.

Profiles of EV charging
At the following figures (fig. 2 to fig.10) are presented the characteristic charging profiles of the considered vehicles in different charge modes and by different values of the initial State Of Charge (SoC) -the maximum charging power is given as a function of the batteries SoC.From those characteristics and the daily grid congestion and PV production profiles, the optimal time for full charge of the vehicles is determined.
The charging stations have a significant impact on the time and mode of charging the EV batteries.The charging devices should provide relatively high instantaneous power, low cost and high efficiency of the power conversion.A classification of charging devices storage station can be made by the following parameters: xlocation of the storage devices: on-board and offboard charger; xdirection of the energy flows: unidirectional and bidirectional charging devices;

Fig. 11 .Fig. 12 .
Fig. 11.Load profile forecast for a sample 24 hour period is the active power in node k; xQk is the reactive power in node k; xVk is the magnitude of the voltage in node k; xVm is the magnitude of the voltage in node m; x (3) is element k, m from the bus admittance matrix; xθk voltage angle at node k; xθm voltage angle at node m; and θkm=θk -θm

Table 1 .
charger placement, starting moment of the charging and the initial SOC are presented in table 1.It is supposed that the most of the EV users will start charging their vehicles after returning home which is considered to occur most probably at 18:30h.The resulting load profile of the grid (fig.13)contains several peaks of the load due to EV charging, although the grid parameters remain in the operational limits, imposed by the operator.The scenario of the operation mode An analytical approach for assessment of the impact of electrical vehicle battery charging on the distribution grid.Based on various EV charge profiles, a sample load curve and PV generators in various nodes of the grid, the time variation of grid parameters is obtained.A set of typical load profiles of EV charging modes is studied and presented.A software implementation and a 24h case study of low voltage distribution network with EV charging devices demonstrate the impacts of EV charging on the low-voltage grid.Based on these results, the interest for future researches on bidirectional EV charging stations providing grid support in the daily periods of peak consumption is proven.This research is funded in the framework of project BG05M2ОP001-2.009-0033 of the EU Structural Funds and "Gestion intelligente des flux énergétiques dans des micro-et nano-réseaux" funded by Agence Universitaire de la Francophonie and the Bulgarian National Fund for Scientific Research.
Fig. 13.Resulting load profile for the sample 24 hour period, including EV charging Conclusion