Impact evaluation of large scale integration of electric vehicles on power grid

Abstract

As the world witnesses a continual increase in the global energy demand, the task of meeting this demand is becoming more difficult due to the limitation in fuel resources as well as the greenhouse gases emitted which accelerate the climate change. As a result, introducing a policy that promotes renewable energy (RE) generation and integration is inevitable for sustainable development. In this endeavor, electrification of the transport sector rises as key point in reducing the accelerating environment degradation, by the deployment of new type of vehicles referred to as PHEV (plug-in hybrid electric vehicle). Besides being able to use two kinds of drives (the conventional internal combustion engine and the electric one) to increase the total efficiency, they come with a grid connection and interaction capability known as the vehicle-to-grid (V2G) that can play a supporting role for the whole power system by providing many ancillary services such as energy storage mean and power quality enhancer. Unfortunately, all these advantages do not come alone. The uncontrolled large scale EV integration may present a real challenge and source of possible failure and instability for the grid. In this work the large scale integration impact of EVs will be investigated in details. The results of power flow analysis and the dynamic response of the grid parameters variation are presented, taking the IEEE 14 bus system as a test grid system.

Graphical abstract

Keywords

PHEV / vehicle-to-grid (V2G) / technical impact / IEEE 14 bus / power flow analysis

1 Introduction

The efforts to reduce global warming are reshaping the way the energy and transport sectors are developing, as they are considered to be main contributors to the increase of greenhouse gases (GHG) concentration in the atmosphere. According to the CO₂ emissions statistics, these two sectors produced nearly two-thirds of global emissions from fuel combustion in 2014, electricity generation being the largest, which accounted for 42%, while transport attributed to 23% of the global emission [1]. Many policies were adopted to promote RE production and integration all over the world, as the cumulative global renewable electricity installed capacity had a growth of 7.9% in 2015 (from 1712 GW to 1848 GW) and the global investment in clean energy increased by 4.1% from 2014 to $329 billion [2].

The electric vehicle (EV) with all types (BEV, PHEV, FCVE) consists a major player in the future sustainable transport systems, and the global deployment of EV is necessary to achieve the GHG emissions reduction and meet the goals of the Paris Declaration on Electro-Mobility and Climate Change that set a global deployment target of at least 20 percent of all road transport vehicles globally to be electrically driven by 2030 [3]. An important milestone was reached in 2015, where more than one million EV were registered to be on road and closing 1.26 million units [4].

EV rises as a solution by introducing the vehicle-to-grid (V2G) concept, and it is the whole concept of using EVs as a distributed resource-load and generation /storage device by their integration into the grid [5], though it is still in the conceptual stages. While EVs are parked which is the most case for vehicles used for commuting, studies have shown that numerous services that have aggregated EVs can bring to the power grid. It can be considered as an auxiliary storage system in a residential distribution system [6,7], providing regulation up and down, therefore, helping the grid operator to ensure the supply-demand equilibrium [8–10], peak load reduction and valley filling [11,12], and frequency regulation services [13,14].

However, the high penetration of EVs can present a real challenge for power grid especially when the charging occurs at peak hours. Several investigation studies were conducted to quantify the resulting impacts from different levels of EVs integration in national grids and IEEE buses test cases [15–18], limiting their influence investigation on the voltage deviation and line losses (steady-state parameters of the grid) where most of problems appear during the transient state of the system, that requires a more detailed study taking the transient parameters in account. In this study, a global influence of EV charging on the power grid will be conducted on the IEEE 14 bus test case.

In the present paper, a general review and modeling of power equation and power flow analysis in the power grid was presented. An IEEE 14 bus system was modeled and simulated under the basic load condition. Besides, the global methodology concerning the power demand EVs and the location of their integration were given and another power flow analysis was conducted. Moreover, the effect of EV integration on the power grid was discussed. Finally, a dynamic study was conducted for voltage and frequency.

2 Power flow and time domain analyses equations of power system

Power flow analysis or known as load flow is a primordial step for power system analysis and design. It is necessary for the planning, operation, scheduling and power exchange between utilities [19]. Besides, it is needed for further analyses such as time domain analysis.

For a power system that consists of N bus, one can write Eq. (1) for complex power injection at each bus [20]

(1){\overline{s}}_n={\overline{v}}_n{\overline{i}}_n^{*}={\overline{v}}_n{\displaystyle \sum _{k\in {\rm B}}{\overline{y}}_{nk}^{*}{\overline{v}}_n^{*},}\hspace{0.17em}\hspace{0.17em}n\in {\rm B}=\{\mathrm{1},\mathrm{2},\mathrm{...},N\},

where {\bar{y}}_{nk} is the element (n, k) of the admittance matrix \bar{Y}and {\overline{y}}_{nk}=g_{nk}+\mathrm{j}{b}_{nk}.

From Eq. (1), the equations of real and reactive power can be developed

(2){\overline{s}}_n=p_n+\mathrm{j}{q}_n={\overline{v}}_n{\displaystyle \sum _{k\in {\rm B}}(g_{nk}-\mathrm{j}{b}_{nk}){\overline{v}}_n^{*},}

where

(3)\{\begin{array}{l}{p}_n=v_n{\displaystyle \sum _{k\in {\rm B}}{v}_n(g_{nk}\cos ({\theta }_{nk})+b_{nk}sin({\theta }_{nk})),\hspace{0.17em}n\in {\rm B}}\\ q_n=v_n{\displaystyle \sum _{k\in {\rm B}}{v}_n(g_{nk}sin({\theta }_{nk})-b_{nk}cos({\theta }_{nk})),\hspace{0.17em}n\in {\rm B}}\end{array}.

Knowing that in power flow equations, the variables are the voltage magnitudes and phases for PQ buses, reactive power, and voltage phase at the generator bus [20]. Equation (3) is nonlinear and has no analytic solution. An iterative technique such as Newton’s method or Jacobi and Gauss-Seidel’s is the most used technique.

Lately, the Newton Raphson method has gained much popularity because of its quadratic convergence. It is most likely to converge even with ill-conditioned problems, and it has been found to be more efficient, practical and the number of iterations required is independent of the system size [19].

The power flow solution includes the voltages and angles at all buses, after setting the slack or the reference bus (whose voltage magnitude and phase angle are predefined) and levels of power demand at PQ buses. One can calculate the real and reactive generated and load levels at all buses as well as the real and reactive flows exchanged through the system lines.

Let x be the vector of unknown parameters of the system Eq. (5), one notices that the right-hand side of Eqs. (3) and (4) depends on the elements of the unknown vector x.

(4)\underline{x}={[{\theta }_{\mathrm{1}}\cdots {\theta }_N,\left|v_{\mathrm{1}}\right|\cdots \left|v_N\right|]}^T.

One can write

(5)\{\begin{array}{l}{P}_n=P_n\left(\underset{\_}{x}\right)\Rightarrow P_n-P_n\left(\underset{\_}{x}\right)=\mathrm{0}\Rightarrow \Delta P_n=\mathrm{0}\\ Q_n=Q_n\left(\underset{\_}{x}\right)\Rightarrow Q_n-Q_n\left(\underset{\_}{x}\right)=\mathrm{0}\Rightarrow \Delta Q_n=\mathrm{0}\end{array},

where P_n and Q_n are the specified injections; while P_n(x) and Q_n(x) are functions of the elements in the unknown vector x, ΔP_n and ΔQ_n are the mismatch power vectors.

The solution update formula is given by

(6){\underset{\_}{x}}^{(i+\mathrm{1})}={\underset{\_}{x}}^{(i)}+\Delta {\underset{\_}{x}}^{(i)}={\underset{\_}{x}}^{(i)}-{\underset{\_}{J}}^{-\mathrm{1}}\left[\begin{array}{l}\Delta \underset{\_}{P}\\ \Delta \underset{\_}{Q}\end{array}\right],

where J is the Jacobian matrix.

The iterative algorithm updates the solution vector x⁽ⁱ⁺¹⁾, until the stopping criteria in Eq. (7) is satisfied.

(7)\left|\Delta P_i\right|\prec {\epsilon }_P;\hspace{0.17em}\hspace{0.17em}\left|\Delta Q_i\right|\prec {\epsilon }_Q.

Time domain analysis or transient stability analysis is a study with the aim of assessing the effect of large disturbances on the power grid. Two approaches are mainly used to conduct this study. The first one which is based on the Lyapunov direct method tries to infer the stability of the ordinary differential equation by building a Lyapunov function able to measure the stability, although it is less time consuming and it has many intrinsic drawbacks such as there is no general systematic method to determine the Lyapunov function [20].

The second solution is to solve the numerical integration. This method is so exact. The so commonly used numerical integration technique is the explicit forward Euler’s method [20].

3 IEEE 14 bus test system at basic load

The IEEE 14 Bus Test Case represents a portion of the American Electric Power System (in the Midwestern US) as of February 1962 [21]. It does not have line limits. Compared to 1990s power systems, it has low base voltages and an overabundance of voltage control capability.

The test system consists of five synchronous machines, three of which are synchronous compensators for reactive power support, 11 loads totaling 259 MW and 81.3 Mvar [21], and lines and transformers. The IEEE 14 bus system is shown in Fig. 1.

The developed model for the IEEE 14 bus system was simulated in the Simulink Simulation Environment. The power flow analysis (PFA) using the Newton-Raphson method was conducted with the following parameters: 100 MVA as base power, 50 Hz for grid frequency, and a power flow tolerance of 10^‒5.

The simulations to determine the voltage magnitude and the voltage angle at each bus were conducted, and the results are depicted in Figs. 2 and 3, respectively.

The load profiles of active and reactive powers are deduced for each bus and plotted in Figs. 4 and 5.

For a more detailed study, the real and reactive power line losses are taken into consideration. The simulation results are shown in Figs. 6 and 7.

4 Effect of EV integration on power grid

To investigate technical impacts of EVs integration on the power grid, a mathematical model is be established first and many parameters should be taken into consideration such as the penetration level, the EV’s battery bank characteristics (capacity and type), and charging level (slow, medium or fast).

Knowing that the PHEV’s charging can be achieved at home using a single-phase domestic socket, at a parking lot at the working place or commercial buildings. The PHEV during its charging requires a constant power input until its battery pack is fully charged [22]. When PHEVs are aggregated in a sizeable number, it can be clustered in one load that sum up them. The new demand is modeled and introduced into the power grid as an additional load tied to distribution feeders.

The new EV loads are distributed throughout the six medium voltage (MV) feeders of the test system (7, 9, 10, 12, 13, and 14) in the network. The vehicle-charging scenario considered in this study is the uncontrolled charging or “dumb charging.” The PHEVs are supposed to start charging as soon as they get home.

An overall penetration of 2070 EVs is assumed, which can be estimated as 0.6 p.u. of real base power demand and about 17% of the total electricity consumption in the distribution system. The EVs are mainly charged at public charging stations and the charging points for vehicle owners in apartment complexes where the parking lots have recharging infrastructure installed so they can be clustered in certain sites. Thus, the EV load demand assembles at several nodes in the grid [5,23,24].

The parameters used for formulating the EV charging load are listed in Table 1. These data are used to approximately estimate the EV battery charging demand and the driving range supported by each home full charging.

A second power flow analysis is conducted on the same model after introducing the estimated demand at the previously mentioned bus feeders. The results of buses voltage magnitude and angles at both base load and PHEV load are depicted in Figs. 8 and 9.

Power analysis results for voltage show a decrease in magnitude for all buses where EVs are integrated because of the growing load demand, except the five buses where the synchronous generators are connected as they provide voltage regulation.

It can be observed from Fig. 10 and Fig. 11 for the real and reactive power profiles of the test system at base load and with PHEV integration that there is a noticeable increase in the generated active power at the first bus to meet the new charging demand as well as the load demand at buses where PHEV are introduced. On the other hand, a constancy in the reactive power demand as the PHEVs is purely modeled as an active power demand.

The power loss analysis presented in Figs. 12 and 13 shows a significant increase in the line active losses as most previous studies showed. This can be easily understood as the increase in the power demand is always associated with a rise in the line current causing more resistive losses. On the contrary, no reactive losses take place as the reactive power cannot be lost or created in the power system, and the reactive power loss is no more than the difference between reactive power values of the sending and receiving end of the transmission line [25].

5 Transient analysis of EVs charging impact

The second step in this study of the resulting impacts is the dynamic or the transient responses of the system variables. A time domain simulation of the IEEE 14 bus gives the following results, as depicted in Fig. 14, for the evolution of the voltages for Bus 12 at base load and with PHEV integration.

In addition to the voltage drop, the oscillation is clearly known to have a bigger amplitude, which can affect the quality of the power at the bus terminal.

The second important parameter in the power grid is the frequency, using frequency measurement tool at the same Bus 12. Figure 15 illustrates the transient and steady-state of the frequency, where it shows a more oscillating behavior and a higher deviation from the base value of the power grid.

6 Conclusions

In this technical study, the solution of power flow analysis applied to the IEEE 14 bus test case, at basic load and with EV integration shows clearly the direct impact of the large-scale integration of EV, especially PHEV in the near future. Voltage drops at the buses where the EVs are integrated and the active power line losses increase due to the increase in the line current.

Besides, it is clearly seen that more oscillating and unstable nature of the curve of voltage and frequency after the EVs are connected, which raises the need to the charging algorithms and study of the allowed rate of penetration of EVs at each specific area, regarding its power generation capacity and power transformers capabilities.

Future work can be oriented to investigating the environmental impacts based on the greenhouse gases emission, and scheduling algorithms to avoid charging the EVs at the peak demand hours, which may cause a failure to the power grid system and prevent extra investments.

References

[1] International Energy Agency. CO₂ emissions from fuel combustion highlights 2016 edition. 2016–01

[2] Beiter P, Tian T. 2015 renewable energy data book. 2015

[3] United Nations Framework Convention on Climate Change (UNFCCC). Paris declaration on electro-mobility and climate change &call to action. 2016–01

[4] International Energy Agency. Global EV outlook 2016 beyond one million electric cars. 2016–01

[5] Guille C, Gross G. A conceptual frame work for the vehicle-to-grid (V2G) implementation. Energy Policy, 2009, 37(11): 4379–4390

[6] Khoucha F, Benbouzid M E H, Amirat Y, Kheloui A. Integrated energy management of a plug-in electric vehicle in residential distribution systems with renewable. In: 2015 IEEE 24th International Symposium on Industrial Electronics, Buzios, Brazil, 2015: 717–722

[7] Tan J, Wang L. Integration of plug-in hybrid electric vehicles into residential distribution grid based on two-layer intelligent optimization. IEEE Transactions on Smart Grid, 2014, 5(4):1774–1784doi:10.1109/TSG.2014.2313617

[8] Karfopoulos E L, Panourgias K A, Hatziargyriou N D. Distributed coordination of electric vehicles providing V2G regulation services. IEEE Transactions on Power Systems, 2016, 31(4): 2834–2846

[9] Lam A Y S, Leung K C, Li V O K. Capacity estimation for vehicle-to-grid frequency regulation services with smart charging mechanism. IEEE Transactions on Smart Grid, 2016, 7(1): 156–166

[10] Lin J, Leung K C, Li V O K. Optimal scheduling with vehicle-to-grid regulation service. IEEE Internet of Things Journal, 2017, 1(6): 556–569

[11] Wang Z, Wang S. Grid power peak shaving and valley filling using vehicle-to-grid systems. IEEE Transactions on Power Delivery, 2013, 28(3): 1822–1829

[12] Sortomme E, El-Sharkawi M A. Optimal scheduling of vehicle-to-grid energy and ancillary services. IEEE Transactions on Smart Grid, 2012, 3(1): 351–359

[13] Han S, Han S, Sezaki K. Development of an optimal vehicle-to-grid aggregator for frequency regulation. IEEE Transactions on Smart Grid, 2010, 1(1): 65–72

[14] Liu H, Hu Z, Song Y, Wang J, Xie X. Vehicle-to-grid control for supplementary frequency regulation considering charging demands. IEEE Transactions on Power Systems, 2015, 30(6): 3110–3119

[15] Axsen J, Kurani K S, Mccarthy R, Yang C. Plug-in hybrid vehicle GHG impacts in California: integrating consumer-informed recharge profiles with an electricity-dispatch model. Energy Policy, 2011, 39(3): 1617–1629

[16] Turker H, Bacha S, Chatroux D. Impact of plug-in hybrid electric vehicles (PHEVs) on the French electric grid. In: 2010 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), Gothenburg, Sweden, 2010

[17] Yilmaz M, Krein P T. Review of the impact of vehicle-to-grid technologies on distribution systems and utility interfaces. IEEE Transactions on Power Electronics, 2013, 28(12): 5673–5689

[18] Dost P, Einwachter F, Spichartz P, Sourkounis C. Influence of electric vehicle charging demands on the grid load based on fleet measurements. In: 2014 IEEE Vehicle Power and Propulsion Conference (VPPC), Coimbra, Portugal, 2014

[19] Saadat H. Power System Analysis. New Delhi: MacGraw-Hill Companies, 1999

[20] Milano F. Power System Modelling and Scripting. Dordrecht: Springer, 2010

[21] Power Systems Test Case Archive Web Site. 14 bus power flow test case. 2017–02–15

[22] Gao S, Chau K T, Chan C C, Wu D Y. Loss analysis of vehicle-to-grid operation. In: 2010 IEEE Vehicle Power and Propulsion Conference, Lille, France, 2011

[23] Gao S. Design analysis and control of vehicle-to-grid services. Dissertation for the Doctoral Degree. Hong Kong: The University of Hong Kong, 2014

[24] Gao S, Chau K T, Liu C H, Wu D Y, Chan C C. Integrated energy management of plug-in electric vehicles in power grid with renewable. IEEE Transactions on Vehicular Technology, 2014, 63(7): 3019–3027

[25] Fetea R, Petroianu A. Reactive power: a strange concept? In: 2nd Europe Conference on Physics Teaching in Engineering Education, 2000

RIGHTS & PERMISSIONS

Higher Education Press