Hot-spot Scenarios of Electric Vehicles on LV Grid including Statistics and Effect of Decentralized Battery Storage

Transcription

1 Hot-spot Scenarios of Electric Vehicles on LV Grid including Statistics and Effect of Decentralized Battery Storage Joel Wenske 1, B. Matthiß 1, J. Binder 1, T. Speidel 2, V. Klaußer 3, M. Klesse 3 1 Center for Solar Energy and Hydrogen Research Baden-Württemberg (ZSW), Germany 2 ads-tec GmbH, Germany 3 Stadtwerke Nürtingen GmbH, Germany - 0 -

2 Center for Solar- and Hydrogen Research Baden-Wuerttemberg (ZSW) Stuttgart: Photovoltaics (with Solab), Energy Policy and Energy Carriers, Central Division Finance, Human Resources and Legal Widderstall: Solar Test Facility ~ 235 employees on 3 locations Ulm: Electrochemical Energy Technologies, Main Building & elab - 1 -

3 R&D Topics at ZSW Energy System Analysis - 2 -

4 Content Motivation 80% of charging at home assumed high simultaneity of charging in residential area Study design Examined Parameters: load and voltage drop Cases: Simultaneous charging Statistics of charging times Effect of stationary battery Conclusions - 3 -

5 Hot-Spots due to Charging Stations in the Low Voltage Grid (LV) Germany: 2017: 0,21% EVs* 2030: up to 21% EVs This study: EVs = s (Battery) Electric Vehicles Load of s + s: PP tttttttttt, ssssssssssssssssssss = PP, ssssssssssssssssssss + PP BBBBBBBB, ssssssssssssssssssss? First Movers : Households () with private Charging Stations (CS) Grid Parameters: Load and voltage drop * datenservice.html#c

6 Limit for Voltage Drop accross the Network HV 110 kv S MV 20 kv S LV 0,4 kv This study: additional voltage drop caused by s is examined UU BBBBBBBB nnnnnnnn = UU + BBBBBBBB nnnnnnnn UU nnnnnnnn DIN EN U N -10% HV/ MV/ LV = High/ Medium/ Low Voltage Grid; S = Transformer Substation U N / U = Nominal/ current Voltage = Household; = Battery Electric Vehicles -5% -5% Assumption: UU BBBBBBBB nnnnnnnn 5% okay UU BBBBBBBB nnnnnnnn 10% critical UU BBBBBBBB nnnnnnnn 10% not permissable - 5 -

7 Example Low Voltage Grid - transformer Load and voltage drop - Low voltage grid of a first mover town district in South Germany Load on 11 transformers (630 kva) Additional voltage drop on last node of 2 overhead power lines (4x50mm², branched structure/ trunk line) - 6 -

8 Study design Cases and Assumptions Cases and Assumptions reduced Peak load/ 2.8 kw 1.4 kw s Penetration - - Charging Station* Charging Pattern Power Probability Daily Arrival Time Daily Driving Distance** Confidence level - - = Household; = Battery electric vehicle; *private charging only; **distriburtion between 0 km km - 7 -

9 Study: Reference Worst Case Cases and Assumptions reduced w/ s CS11 w/ CS14 w/ CS22 Peak load/ 2.8 kw 1.4 kw 2.8 kw s Penetration % Charging Station* Charging Pattern Power - - All at once 11 kw All a. o kw All a. o. 22 kw Probability Daily Arrival Time Daily Driving Distance Confidence level = Household; = Battery electric vehicle; *private charging only - 8 -

10 Results for Peak Load of all connected Housholds (Reference Case) Load in [kva] Transformer no. Load in [kva] 1400 Peak load () reduced Peak load () 1200 Peak load () + charging load (@ 11 kw CS, ) 99.7%) 1000 Peak load () + charging load (@ 14.7 kw CS, ) 99.7%) Peak load () + charging load (@ 22 kw CS, ) 99.7%) 800 permissible transformer load Transformer no. = Household CS = Charging Substation Conservative Design: 400 Transformers are save for all connected Households at Peak Loads - 9 -

11 Results: Overload on transformers if all 21% s charge simulanously Load in [kva] Transformer no. Load in [kva] 1400 Peak load () reduced Peak load () 1200 Peak load () + charging load (@ 11 kw CS, ) 99.7%) 1000 Peak load () + charging load (@ 14.7 kw CS, ) 99.7%) Peak load () + charging load (@ 22 kw CS, ) 99.7%) 800 permissible transformer load 11 kw CS: Load 400doubled vs. peak load Load tripled vs. reduced 200 = Household CS = Charging Substation Permissable 0 Transformer load is exceeded in nearly 2all cases Transformer no

12 Study: Reference Worst Case Cases and Assumptions reduced w/ s CS11 w/ CS14 w/ CS22 Peak load/ 2.8 kw 1.4 kw 2.8 kw s Penetration % Charging Station* Charging Pattern Power - - All at once 11 kw All a. o kw All a. o. 22 kw Probability Daily Arrival Time Daily Driving Distance Confidence level = Household; = Battery electric vehicle; *private charging only

13 Study: Reference Worst Case Probabilistic View Cases and Assumptions reduced w/ s CS11 w/ CS14 Peak load/ 2.8 kw 1.4 kw 2.8 kw s Penetration % Charging Station* Charging Pattern Power Probability Daily Arrival Time Daily Driving Distance - - All at once 11 kw Confidence level All a. o kw = Household; = Battery electric vehicle; *private charging only w/ CS22 All a. o. 22 kw likelihood of occurrence of arrival time in [%] 0,2 0,2 0,1 0,1 0,0 Maximum Driving Distance (km) Probability (%) time of day [h] 0 29,9 1 2, , ,7 Adjusted from Probst, M. Braun, S. Tenbohlen, Erstellung und Simulation probabilistischer Lastmodelle von Haus-halten und Elektrofahrzeugen zur Spannungsbandanalyse, IEH Universität Stuttgart, Germany: Internationaler ETG- Kongress, Würzburg, Beitrag Nr. P06, 2011; Adjusted from R.Follmer et al.., Mobilität in Deutschland 2008, infas, DLR and, Bundesministerium für Verkehr, Bau und Stadtentwicklung, 2010, Bonn, Germany 40 14,7-12 -

14 Study: Reference Worst Case Probabilistic View Cases and Assumptions reduced w/ s CS11 w/ CS14 w/ CS22 w/ CS11 w/ prob w/ CS14 w/ prob w/ CS22 w/ prob Peak load/ 2.8 kw 1.4 kw 2.8 kw 2.8 kw s Penetration % 21% Charging Station* Charging Pattern Power - - All at once 11 kw All a. o kw All a. o. 22 kw Daily after arrival 11 kw D. a. arrival 14.7 kw D. a. arrival 22 kw Probability Daily Arrival Time Daily Driving Distance** max. 6 p.m., = 50 km (0 km km) Confidence level % = Household; = Battery electric vehicle; *private charging only; **distribution 30% do not drive daily, rest = 50 km

15 Study: Reference Worst Case Probabilistic View Cases and Assumptions reduced w/ s CS11 w/ CS14 w/ CS22 w/ CS11 w/ prob w/ CS14 w/ prob w/ CS22 w/ prob Peak load/ 2.8 kw 1.4 kw 2.8 kw 2.8 kw s Penetration % 21% Charging Station* Charging Pattern Power - - All at once 11 kw All a. o kw All a. o. 22 kw Daily after arrival 11 kw D. a. arrival 14.7 kw D. a. arrival 22 kw Probability Daily Arrival Time Daily Driving Distance** max. 6 p.m., = 50 km (0 km km) Confidence level %

16 Results: No overload on transformers for 21% s if Probabilistic View included Load in [kva] With daily driving distance, arrival time and 99.7% confidence level: 800 Reduced 600 Number of s charge simultaneously 400 Permissable transformer load is 200 (mostly) not reached Transformer no. Load in [kva] 1400 Peak load () reduced Peak load () 1200 Peak load () + charging load (@ 11 kw CS, 99.7%) 1000 Peak load () + charging load (@ 14.7 kw CS, 99.7%) Peak load () + charging load (@ 22 kw CS, 99.7%) 800 permissible transformer load Transformer no

17 Study design Additional Voltage Drop Cases and Assumptions reduced w/ s CS11 String A/B w/ CS14 w/ CS22 w/ CS11 w/ prob String A/B w/ CS14 w/ prob w/ CS22 w/ prob s Penetration 5/ 10/ 21% 5/ 10/ 21% Charging Station* Charging Pattern Power Distribution on string Probability All at once 11kW first/ last node or equally distributed All at once Daily after arrival 11kW first/ last node or equally distributed 6 p.m. = 50 km Confidence level irr. 99.7% = Household; = Battery electric vehicle; *private charging only Adjusted from

18 Results: Voltage drop inadmissible for >10% s local distributed If all s charge simultaneously (Worst Case) Excessive additional voltage drop UU for >10% s voltage drop [p.u.] 0% -2% -4% -6% -8% -10% -12% -14% -16% -18% Total No. of s CS at 1st node CS equally distributed 4 s (5%) 8 s (10%) 17 s (21%) String A (60 ), 11kW CS CS at last node 2 s (5%) 4 s (10%) 8 s (21%) - String B (30 ), 11kW CS

19 Study: Reference Worst Case Probabilistic View Cases and Assumptions reduced w/ s CS11 String A/B w/ CS14 w/ CS22 w/ CS11 w/ prob String A/B w/ CS14 w/ prob w/ CS22 w/ prob Peak load/ 2.8 kw 2.8 kw s Penetration 5 /10 / 21% 5 /10 / 21% Charging Station* Charging Pattern Power Probability Daily Arrival Time Daily Driving Distance** Battery Supply (Charging firstly by Battery than by Grid supply) All at once 11 kw - Probabilistic View w/ Battery Supply Daily after arrival 11 kw Max. 6 p.m., = 50 km** 5 kwh 10 kwh 15 kwh 20 kwh

20 Total number s in the low voltage string number of s (simultaneously charging) % 10% 21% 100% 5% 10% 21% 100% penetration String A penetration String B Total number of

21 Total number vs. statics of simultaneous charging (99,7% confidence) number of s (simultaneously charging) % 10% 21% 100% 5% 10% 21% 100% penetration String A penetration String B Total number of Remaining s w/ probalistic view (@6 p.m., 99.7% confidence)

22 Reduction of number simultaneous charging s 18 on network supply due to 5kWh Batteries Only 34% of s remain totally number of s (simultaneously charging) % 10% 21% 100% 5% 10% 21% 100% penetration String A penetration String B Total number of simulaneously charging s Remaining s w/ probalistic view Remaining s w/ prob. view + 5kWh battery

23 Reduction of number simultaneous charging s 18 on network supply due to Batteries number of s (simultaneously charging) % 10% 21% 100% 5% 10% 21% 100% penetration String A penetration String B Total number of simulaneously charging s Remaining s w/ probalistic view Remaining s w/ prob. view + 5kWh battery Remaining s w/ prob. view + 10kWh battery Remaining s w/ prob. view + 15kWh battery Remaining s w/ prob. view + 20kWh battery

24 Summary and Outlook Simultaneous charging of all existing s leads to excessive transformer loading and voltage drop on string in LV grid even for low numbers of s. With the statistics of arrival times and driven distance over the day not more than 7 out of 17 s in this string will charge simultaneously at 99.7 % confidence level Small decentral batteries of 5 kwh at each CS further reduce significantly the maximum number of vehicles to be expected for simultaneous charging from the grid. Outlook: It will not pay to extend the grid for low probabilities of high demand. Therefore, communication to limit charging for the unlikely event of a high number of simultaneous charging requests seems the most economical option

25 // Energy with a future // Energy with a future // Center for Solar Energy and Hydrogen Research Baden-Württemberg // Zentrum für Sonnenenergie- (ZSW) und Wasserstoff- Forschung Baden-Württemberg (ZSW) // Joel Wenske M.Sc. joel.wenske@zsw-bw.de Thank you for your attention! Stuttgart: Photovoltaics, Energy Policy and Energy Carriers, Central Division Finance, Human Resources & Legal Widderstall: Solar Test Facility Ulm: Electrochemical Energy Technologies, Main Building & elab