Impact of Electric Vehicles on Voltage Profile and ... - IEEE Xplore

12 downloads 81404 Views 341KB Size Report
may negatively affect distribution network with consequent power quality issues including distorted voltage profile, har- monic contamination, increased power ...
Impact of Electric Vehicles on Voltage Profile and Harmonics in a Distribution Network Azhar Ul-Haq, Carlo Cecati DISIM, University of L’Aquila 67100 L’Aquila, Italy [given_name.family_name]@univaq.it

Abstract—This paper focuses on potential power quality impact of electric vehicles in terms of voltage profile and harmonic contamination in a low voltage CIGRE benchmark distribution system. Voltage drop and harmonic distortion have been amongst the main power quality concerns in view of likelihood of largescale penetration of electric vehicles in transportation system. Simulations are carried out to quantify the power quality index at different electric vehicle penetration levels while considering three distinct charging strategies. The power quality indexes are found much beyond their allowed limits at only 45 % EV penetration.

I. I NTRODUCTION Electric vehicles (EVs) require certain amount of electrical energy from power grid [1], [2] and their large-scale penetration may negatively affect distribution network with consequent power quality issues including distorted voltage profile, harmonic contamination, increased power flow, transformer overloading, and others negative effects [4], [5]. Recently, some considerations have been made to discuss limitations of power system for EVs charging [6]-[11]. Power quality parameters such as voltage profile, harmonic contamination, frequency variations, voltage drop/dip [12] and their indexes are restricted by a number of relevant standards. Therefore, it becomes quite important to identify and determine the presence of voltage drop and harmonics in a circuit in order to deal with it timely for trouble free operation of power system. This paper aims at summarizing and determining power quality indices associated with large number of EV charging. Focus of this paper is a sample residential low-voltage (LV) distribution network that is based on the CIGRE low voltage European benchmark configuration. The distribution network is tested and analyzed while considering three distinct EV charging strategies including uncontrolled charging scheme, tariff based charging and EV controlled charging strategy at different penetration levels of EVs. Voltage total harmonic distortion (THD) that results from EV penetration with a number of harmonic spectrum is studied at current THD of 17.4 %. The study is performed while taking into account different scenarios of light and heavy loads. The paper is organized as follows; problem formulation is explained in Section II. Types of EV charging schemes are described in Section III . Explanation of the system under study,

Abbasi Ehsan, Kai Strunz School of El. Engr., Technical University of Berlin Berlin 10587, Germany [given_name.family_name]@tu-berlin.de

network description and simulations, assessment of harmonic distortion in the system and analysis of the obtained results are given in Section IV, IV-A and Section IV-A1 respectively. The paper is concluded in Section V. II. P ROBLEM F ORMULATION This section formulates the problem of voltage drop and harmonic distortion caused by charging EVs in a distribution network. The system is considered with high amount of load for demonstrating a worst voltage drop scenario. Typically, an EV charging is done via two power converters including ACDC power converter that exchanges power between charger and the grid and DC/DC power converter that controls battery charging. Harmonics caused by EV charging are mainly related to interfacing circuit topology connecting to the network [14]. A. Estimating Voltage Profile and Evaluating Harmonics A certain amount of EV load could be connected to the distribution system until it starts affecting power quality parameters. Voltage drop ∆VD from point Rk to the end of distribution feeder can be expressed as in Eq. (1) p P2 + Q 2 z1 ∆VD = Rk (n − Rk + 1) (1) Vnom where, z1 represents impedance of the line segments (of same length) between two nodes of the network, respectively, n is the total number of nodes of the system, P and Q are active and reactive power and Vnom is the nominal voltage. Voltage drop from source point to an arbitrary node ∆Vl b can be found as given in: p Rk P2 + Q 2 z1 ∆Vl b = (Rk − 1) (2) 2 Vnom then, the total voltage drop at a node Rk would be: p Rk P2 + Q 2 z1 ∆VD = (2n − Rk + 1) . 2 Vnom

(3)

Voltage at point Rk in the distribution network, before addition of EVs charging load may be calculated as:

978-1-4673-6765-3/15/$31.00 ©2015 IEEE

∆VR k = Vo −

Rk (2n − Rk + 1) 2

p

P2 + Q 2 z1 Vnom

(4)

Table I L IMITS FOR CURRENT HARMONIC SET BY IEC 61000-3-2 FOR EQUIPMENT WITH A MAXIMUM INPUT CURRENT OF 16 A. Order 2, 3 4, 5 6, 7 9 15≤ n ≤ 39 8≤ n ≤ 40

voltage drop after addition of EVs can be calculated using Eq. (5)

∆VEV = Rk

p

P2 + Q 2 z1 Vnom

(5)

net voltage drop ∆VT D k can be obtained subtracting (5) from (3):

∆VT D k = Vo −

Rk 2

−Rk

(2n − Rk + 1) √



P 2 +Q 2 z 1 Vno m

(6)

P 2 +Q 2 z 1 . Vno m

T H D χ, =

h=2

χ hR

χ1R

The same standard define voltage harmonics limits, in particular, according to EN 50160, THD in an LV or MV system should not be more than 8% under normal operation; IEC 61000-2-2 also states that voltage THD at the buses carrying up to 69 kV must remain below 5%. As this work deals with single phase EV charging so charging current for an EV charging should not be considered beyond 16 A per phase. III. EV C HARGING S TRATEGIES

From Eq. (6), it could be observed that the voltage drop after accession of EVs to the power grid is significantly enhanced. The IEC 61000 and EN 56160 are found to be the most relevant standards, the latest defines that during a period of week, the supply voltage must not violate the limit of ±10 of the nominal value and the supply voltage should not outbreak a range of -15% to +10% for a period of 10 minutes [15] and the first also limits voltage drop within ±10% [16]. It is important to note that this allowed limit of voltage drop is accounted for on the whole voltage drop i.e. it is overall voltage drop on medium voltage network MV/LV transformer, LV feeder, cables and connections. For harmonic evaluation the system modeling is carried out with harmonic current sources. The relevant admittance matrix is changed in accordance with [17]. A model of resistive load in parallel with reactance is utilized in [18]. Total current and voltage harmonic distortion expressed after connection of EV charging load under UEVC, TEVC and CEVC methods are given by: qP  H

Harmonics 1.08, 2.3 0.43, 1.14 0.30, 0.77 0.40 0.15 ∗ 15/n 0.23 ∗ 8/n

2

(7)

where h represents harmonic order, H is the highest number of harmonic, χ hR i.e. either VRh or I Rh are the RMS values of the h-th component and χ1R is the fundamental frequency component. Standard IEC 61000, part 3-2, states allowable limits for current harmonics for electrical equipment with maximum input phase current of 16 A connected to a public power distribution systems. Part 3-12 deals with harmonics content for electrical equipment having input current of 16 A to 75 A per phase. Table I contains the allowable harmonics for each order [19].

Three different EV charging strategies including uncontrolled EV charging, tariff based EV charging and controlled EV charging are considered in this paper, those are illustrated here below. A. Uncontrolled EV Charging In an uncontrolled EV charging (UEVC) strategy, EV charging can be freely done on “whenever and whenever needed” basis without any incentives and penalty for charging and restrictions respectively. Maximum number of EVs that can be charged simultaneously at all the available charging points in a power distribution system can be formulated as: EVs number being charged at hour h can be found as given in Eq. (8):

EVN r (ma x) =

24  X

t (h)=1

 S P EVCHt (h) + EVCWt (h) + EVCPtP(h) ∗ EVP L (8)

EVN r (ma x) = Maximum number of EVs to be recharged at hour h S EVCHt (h) = EV charging at home at hour h P EVCWt (h) = EV charging at work place/commercial center at hour h EVCPtP(h) =EV charging at public place at hour h

EVP L = EV penetration level. Sum of the power drawn from an LV distribution network for recharging all ranges of EVs can be expressed as in Eq. (9) and power required to charge EVN r (ma x) can be calculated using Eq. (10):

PEVN r (ma x) = EVN r (ma x) ∗ PC (h) EV

(9)

where, PC (h) EV represents power for recharging an EV, P putting, EVN r (ma x) and EVCWt (h) = 0, Eq. (9) becomes:

PEVN r (ma x) =

24 (  X

t (h)=1

MV Distribution System: 20kV Line-to-Line

R0

 ) S EVCHt (h) + EVCPtP(h) ∗ PC (h) EV ∗ EVP L

Load Bus Supply Point

(10) R1

35m

PEVN r (ma x) = Active power required to charge EVN r (ma x) .

Ground/Earthing Plate

R2

B. Tariff Based EV Charging In tariff based EV charging (TEVC) strategy price of electricity is not considered fixed at all times and EVs can be preferably recharged during valley hours. In this charging strategy amount of power drawn at hour h could be either less than the required total charging power (as compared with uncontrolled EV charging) or equal to that. High price tariff is considered during peak hours between 13:00 to 20:00 hours of a day [20]. Amount of power required at hour h to recharge all ranges of EVs at a residential unit and public points is given in Eq. (11 and 12) and total power drawn from an LV distribution network for charging EVN r (ma x) can be expressed as given in Eq. (13).

R11 30m

LV Distribution System: 400V Line-to-Line

R3 30m

R4

R12

R13

30m 30m

R15 R5

R14

35m

R16 R6

R7 35m

R8

PEV H S ≤ C t (h)

PEV P P ≤ C t (h)

PEVN r (ma x) ≤

24  X

t (h)=1

24  X

t (h)=1

24 (  X

t (h)=1



S EVCHt (h) ∗ PC (h) EV ∗ EVP L

EVCPtP(h) ∗ PC (h) EV ∗ EVP L

(12)

t (h)=1



B17

30m 35m

30m

R10

Figure 1. Configuration of urban residential distribution systems R15 and R16 after connection of EV charging load under UEVC, TEVC and CEVC methods.

(13)

C. Controlled EV Charging The controlled EV (CEVC) charging is meant to manage EV load demand to avoid stress on the power grid. The upper limit is considered with perspective of daily load curve [20]. In order to prevent overloading of feeder, EV charging is allowed when maximum power demand on a load node is less than the threshold value otherwise EV drivers are directed to the nearest available charging slot. At hour h, total power demand including EV load demand, as given in Eq. (14) is compared with maximum permissible power supply from a node as given in Eq. (15), while the charging request is accepted if total demand is less than maximum allowable value.

P24

R9 R18



 ) S EVCHt (h) + EVCPtP(h) ∗ PC (h) EV ∗ EVP L

PN L o =

35m

(11)

P24

t (h)=1 PN L t (h)

  S EVCHt (h) + EVCPtP(h) ∗ PC (h) EV ∗ EVP L

(14)

if PN L o < PN L ma x

(15)

otherwise, PC (h) EV = 0 where, PN L o stands for total power demand at a node load R after accession of EVs, PN L ma x represents maximum power demand that can be fed at hour h, and PC (h) EV is the amount of power for recharging EVs. IV. S YSTEM D ESCRIPTION AND S IMULATIONS The benchmark LV distribution network represents a real world LV distribution system which is more user friendly and flexible. In the considered network by CIGRE Task Force C6.04.02 [20], addition of EVs load to the whole power demand is specified in Table II. An MV/LV transformer supplies a maximum load of 179 kW which is almost 60% of the rated capacity. The urban distribution topology is shown in Fig. 1. Amount of voltage drop at different load buses before addition of EV charging load is shown in Fig. 2, that depicts that the

Table II L OAD PARAMETERS OF URBAN DISTRIBUTION SYSTEM . Number of Households 9 25 48 09 21

Peak Load (kVA) 16 33 68 14 45

1.06 Uncontrolled Charging Tariff Based Charging Controlled Charging

1.02

voltage drop at the load buses before addition of EV charging load is within normal limits. Addition of EV charging load on each node is kept proportional to the previously connected load and additional load is dispersed on all the three phases. Voltage drop on the load buses is observed with EV penetration levels of 30%, 45% and 60% with connection of 17, 26 and 34 EVs respectively. Voltage drop on the load buses is shown in Fig. 3.

Voltage drop (p.u)

Node R11 R15 R16 R17 R18

0.98

0.94

0.9

4

8

12 Time (hours)

16

20

24

Figure 4. Voltage profile at load bus R15 with EV charging load

7

The presented values show that the voltage drop becomes significant at EV penetration level of 45%. It may be observed that the load buses located far from the main supply point such as R18 are more susceptible to occurrence of voltage drop after accession of EVs in the system. Voltage profile recorded under all three EV charging schemes is depicted in Fig. 4.

Voltage drop (%)

5.25

3.5

1.75

0

A. Simulations for Assessment of Harmonic Distortion

R11

R15 R16 R17 Load buses (number)

R18

Figure 2. Voltage drop on load buses before connection of EVs.

10 R15: UEVC R15: TEVC R15: CEVC

Voltage drop (%)

8.5

R16: UEVC R16: TEVC R16: CEVC

7

5.5

4 0

15

30 45 EV penetration level (%)

60

Figure 3. Voltage drop on load buses R15 and R16 after connection of EV charging load under UEVC, TEVC and CEVC methods.

Voltage THD that may result from EV penetration with a number of harmonic spectrum is studied. The study is performed while taking into account different scenarios related to loads that generate harmonic in the system. Harmonics up to 25t h order are taken into account here, as the higher order harmonics are not any significant. Two scenarios including light and heavy loads are considered and different groups of loads are considered for each scenario. Characteristics of the considered loads are based on the data given in [21], [22]. Current harmonic distortion produced by some of loads such as electric dryers and heaters can be neglected. On the other hand most of daily-used appliances including florescent lamps (FLs), PCs, and monitors inject total current harmonic distortion of amount of 101.4%, 99.3% and 93.5% respectively, into the power grid [21]. 1) Analysis: Simulations are carried out to quantify the resulting voltage THD at different nodes of the system for the mentioned light and heavy load scenarios at EV penetration levels of 45%, 65% and 95%. Where, two worst cases are considered in terms of phase angles of EV charging current harmonics; i) a case of EV current harmonics to be in phase and ii) in opposite phase of the harmonic currents that result from other sources.

Table III VOLTAGE THD

IN DISTRIBUTION FEEDER WITH CURRENT THD OF FOR TWO CASES OF CURRENT HARMONIC PHASE ANGLES .

Loading scenario light load heavy load

Harmonics EV PL: 45% I.P./O.P. 5.16/2.33 8.11/3.13

Harmonics EV PL: 65% I.P./O.P. 7.89/5.06 9.76/4.18

17.4%,

Harmonics EV PL: 95% I.P./O.P. 9.93/8.02 14.24/7.19

It may be observed that the obtained THD in the system depends on the phase difference of harmonics generated from other loads and the harmonics caused by EV charger and this difference may be significant at high EV penetration levels. V. C ONCLUSION

Figure 5. Voltage THD in distribution feeder with current THD of 17.4% under light loading.

It becomes a visible fact that a massive adoption of electromobility will significantly alter the existing power system operations and practices. Two important power quality parameters including voltage profile and harmonic distortion at different EV penetration levels are determined and quantified in this paper. The presented research approach helps identify a penetration level of EVs that can be fed with power supply safely i.e. percentage of EVs that does not pose any negative impact on the system. The distribution network under study is tested and analyzed under three different EV charging strategies including uncontrolled charging, tariff based charging and controlled EV charging. The uncontrolled EV charging exhibits worst results whereas some betterment (less negative impact on the grid) is observed under controlled EV charging scheme. The obtained results show that an EV penetration level of 45% can cause significant voltage drop in the system and voltage THD with a current THD of 17.4% is found to be beyond allowable limit of 8% even under light loading scenario. In case of considered heavy load scenario, voltage THD can become as high as 14.2%. Therefore, certain measures and upgradation requirements must be considered before addition of large number of EVs in the system.

Figure 6. Voltage THD in distribution feeder with current THD of 17.4% with heavy load.

R EFERENCES

As aforementioned, values of even harmonics and its order of above 25t h harmonics are neglected here. The related harmonic spectrum can be found in [23]. Simulation results for current THD of 17.4% with in phase (I.P.) and in opposite phase (O.P.) angle of EV charger current harmonics are given in Table III. The results shown in Fig. 5 and Fig. 6 illustrate that the appeared voltage THD in the distribution system goes beyond allowed limit of 8% in many cases. Thus, it could be readily established that EVs could significantly contaminate the distribution system with high amount of harmonic injection. For the worst considered scenario of harmonics phase angles, the voltage harmonic distortion may touch the limits even at EV penetration level of 45% or above.

[1] D. Wu, D.C. Aliprantis, K. Gkritza, "Electric Energy and Power Consumption by Light-Duty Plug-In Electric Vehicles," IEEE Trans. Power Sys., vol. 26, no. 2, pp. 738-746, May, 2011. [2] Gan Li, Xiao-Ping Zhang, "Modeling of Plug-in Hybrid Electric Vehicle Charging Demand in Probabilistic Power Flow Calculations," IEEE Trans. Smart Grids, vol. 3, no. 1, pp. 492-499, Mar. 2012. [3] K. Clement, K. Van Reusel, and J. Driesen, “The consumption of electrical energy of plug-in hybrid electric vehicles in Belgium,” in Proc. 2nd Eur. Ele-Drive Transportation Conf., Brussels, May 2007. [4] Rui Shi, “The Dynamic Impacts of Electric Vehicle Integration on the Electricity Distribution Grid,” M.Phil thesis, Sch. of Electron. Elect. and Comp. Engg, The Univ. of Birmingham, UK, Nov. 2012. [5] L. P. Fernández, T. G. S. Román, R. Cossent, C. M. Domingo, and P. FrÃas "Assessment of the impact of plug-in electric vehicles on distribution networks," IEEE Trans. Power Sys., vol. 26, no. 1, pp. 206 -213, Feb. 2011. [6] L. Zhao, S. Prousch, M. Hubner, A. Moser, "Impact Assessment of Varying Penetrations of Electric Vehicles on Low Voltage Distribution Systems," in Proc. of IEEE PES Transmission and Dist. Conf. and Expo., pp. 1-6, Jul. 2010.

[7]

[8] [9]

[10]

[11] [12] [14] [15]

V. Tikka, J. Lassila, J. Haakana, J. Partanen, "Case Study of the Effects of Electric Vehicle Charging on Grid Loads in an Urban Area," in Proc. of IEEE PES Int. Conf. and Exhibition on Innovative Smart Grid Techno., pp. 1-7, Dec. 2011. Z. Darabi, M. and Ferdowsi, "Aggregated Impact of Plug-in Hybrid Electric Vehicles on Electricity Demand Profile," IEEE Transactions on Sustainable Energy, , vol.2, no.4, pp.501-508, Oct. 2011. E. Sortomme, M. M. Hindi, S. D. J. MacPherson, and S. S. Venkata, “Coordinated charging of plug-in hybrid electric vehicles to minimize distribution system losses,” IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 198–205, Mar. 2011. J.A. Orr, A.E. Emanuel, D.J. Pileggi, "Current Harmonics, Voltage Distortion, and Powers Associated with Electric Vehicle Battery Chargers Distributed on the Residential Power System," IEEE Trans. Ind. App., vol. Ia-20, no. 4, pp. 727 – 734. Aug. 1984. K. Clement-Nyns, E. Haesen, and J. Driesen, “The impact of charging plug-in hybrid electric vehicles on a residential distribution grid,” IEEE Trans. Power Syst., vol. 25, no. 1, pp. 371–380, Feb. 2010. R.C. Dugan, M.C. McGranaghan, and H.W. Beaty, Electrical Power Systems Quality, New York, McGraw-Hill, 1996. R. Bass, R. Harley, F. Lambert, V. Rajasekaran and J. Pierce, “Residential harmonic loads and ev charging,” in Proc. of IEEE Power Engineering Society Winter Meeting, vol. 2, pp. 803–808, Feb. 2001. EN 50160, “Voltage characteristics of electri city supplied by public distribution systems, ” September 2007.

[16] H. Markiewicz, A. Klajn, “Voltage Disturbances: EN 50160, Voltage characteristics of electricity supplied by public distribution systems, ” Leonardo Power Quality Initiative, Copper Development Association, July 2004. [17] Sun. Yuanyuan, Z. Guibin, Xu Wilsun, J.G. Mayordomo, "A Harmonically Coupled Admittance Matrix Model for AC/DC Converters," IEEE Trans. Power Syst., vol. 22, no. 4, pp.1574-1582, Nov. 2007 [18] M.A.S. Masoum, M. Ladjevardi, A. Jafarian, E.F. Fuchs, "Optimal placement, replacement and sizing of capacitor Banks in distorted distribution networks by genetic algorithms," IEEE Trans. Power Deliv., vol. 19, no. 4, pp. 1794-1801, Oct. 2004 [19] M. Nazarudin Zainal Abidin, “IEC 61000-3-2 Harmonics Standards Overview,” Schaffner EMC Inc., Edsion, NJ, USA, May 2006 [20] K. Strunz, N. Hatziargyriou, C. Andrieu "Benchmark Systems for Network Integration of Renewable and Distributed Energy Resources," CIGRE Task Force C6.04.02, 2013 [21] J. Cunill-Sola, M. Salichs, "Study and Characterization of Waveforms From Low-Watt (