Fleets of electric vehicles as adjustable loads Facilitating the integration of electricity generation by renewable energy sources

Transcription

1 Fleets of electric vehicles as adjustable loads Facilitating the integration of electricity generation by renewable energy sources Katrin Seddig 1,2 ; Patrick Jochem 2 ; Wolf Fichtner 2 1 Energy Solution Center (EnSoC) 2 Karlsruhe Institute of Technology (KIT), Institute for Industrial Production (IIP), Chair of Energy Economics New York, th IAEE International Conference

2 Agenda Introduction Methods and data Results Conclusion and outlook 2

3 Introduction Situation Energy Transition in Germany aims to reduce greenhouse gas emissions, e. g. by Energy efficiency increase in industry Build-up of wind energy and photovoltaic systems Growing share of electric vehicles (EV) and alternative fuels in the transportation sector Complication An Increasing number of EV cause Increase of energy consumption of a representative household Influence of the grid load (i.e. peak power) Objections of the actual research project ( Integrated fleet and charge management ) Use of fleet of EVs as controllable loads (i.e. controllable electricity consumption in time and power) Quantification of their load shifting potential Integration of renewable energy sources (RES) Impacts of different charging strategies 3

4 Electricity Supply [MWh] Installed Capacity in GW Share of RES [%] Increased share of renewable energy sources in Germany Development of electricity consumption from RES in German from Geothermal energy Photovoltaics Biomass Wind energy Hydropower Years Data source: BMWi (2013) Development of electricity supply of installed capacity and from PV plants in Germany Electricity supply Installed capacitiy Years Data source: BMWi (2013), Fraunhofer ISE (2014) 4

5 Demand Response Corresponding load shifting potentials might equalize the supply side fluctuations and could therefore help to increase the share of electricity generation by RES Possibilities to integrate EVs for load shifting Avoid peak-loads (peak shaving) Avoid low loads (valley filling) Load shifting Source: Based on Jochem et al. (2013) Might lead to a stabilization of the grid and it could be used as a balancing capacity 5

6 Multi-agent based simulation - Overview of main agents Agents Electric Vehicles Fleets of EV Charging points Specifications: - Battery capacity : 24 kwh - Average electricity consumption: 0.2 kwh/km Specifications: - Short term parker - Long term parker - Commercial fleet Charging power: kw - 22 kw Assumptions, e. g.: For each electric vehicle is a charging point available and for the duration of the parking time, it is connected Each EV is assigned to a mode of charging power, however within this, it is possible to charge with varying charging power 6

7 Model structure Calculations are performed in a 1-minute resolution (t {1,, T}) Limitation of the power input for the parking garage is foreseen It depends on the maximum power of the building connection (P BC ) As long as there is enough charging power available for the EV (P EVSE ) at the charging points i {1,, I}, the EV charges at the residual load P BC i P EVSEi,t t; with P BC Maximum power of building connection P EVSE Charging power of EV at charging point i EVSE Electric vehicle supply equipment If maximum power is reached no new customer request for charging will be accepted, until charging power is available (due to a finished charging process from another EV) If a charging request of an EV is denied the success rate drops If parking time is longer than charging time, then load shifting is possible 7

8 Relative frequency [%] Relative frequency [%] Input data for each fleet of the electric vehicles For commercial fleets data was statistically analyzed to derive synthetical driving profiles Short and long term parker were analyzed using data from a parking garage Parking duration of EV fleets during a weekday in a parking garage Distance of trip (national wide data) Short term parker Commuter Short term parker (Leisure, Shopping, others) Commuter Length of stay [h] Distance [km] Data source: PBW (2013) Data source: BMVBS (2012) 8

9 Overview of chosen scenarios Uncoordinated charging (Mode 2 or Mode 3) Summer PV Winter PV Without PV Strategy SoC 100 x x Strategy SoC next trip For all scenarios 75 commercial EVs, 150 long term and 150 short term EVs are considered PV specifications: On the roof of the parking garage maximum usable potential of 200 kw p Summer PV: total feed-in of 8420 kwh (Mo - Fr) Winter PV: electricity supply of 333 kwh (Mo - Fr) 9

10 Impacts of the charging strategies Uncontrolled charging without PV Maximum Power of BC (P BC ) Accumulated charging power of all fleets Commercial fleet Short term parker Long term parker Without PV the success rate for the requests of EVs drops As maximum power limit of the building connection is reached 10

11 Impacts of the charging strategies Uncontrolled charging with summer feed-in data for PV Available charging power Accumulated charging power of all fleets Commercial fleet Short term parker Long term parker Charging curve changes Higher flexibility for load shifting is available Nearly no charging over night is necessary, however, possible (especially for the commercial fleets) 11

12 Impacts of the charging strategies Using preferable PV Available charging power Used PV energy for charging Accumulated charging power of all fleets Other electricity used for charging With adapted charging scheduling algorithm: 82 % of the electricity could be supplied by the PV plants Up to 100% of generated PV electricity can be used 12

13 Impacts of the charging strategies CO 2 emission reduction by each fleet Available PV charging power Short term parker Long term parker could reduce the most CO 2 emissions Commercial fleet Long term parker 1.23 t CO 2 reduction (if German average CO 2 emission factor for the energy mix of 477 g CO 2 /kwh el is used) Have the highest load shifting potential during a day due to their long parking time 13

14 Conclusion An integration of (local) electricity generation by RES is promising Leads to a release of the local electricity grid Generated PV electricity of the parking garage can be used nearly up to 100 % Charging behavior of commuters can be adapted towards PV integration Results of the simulation show: Load shift potential towards different parts of the day is significant and might substitute local storages or grid extensions Commercial fleets might provide high load shift potentials during nights e. g. for wind integration 14

15 Outlook Implementation of monetary values Adaption of charging strategies Priorization of fleets depending on the willingness to pay (different scheduling algorithms) Developing different business models for the involved parties (e. g. energy supplier, parking garage operator) Integration of wind energy as one part of the charging power Could improve the flexibility for load shifting Improvement of the profitability of public charging stations through a charging scheduling algorithm 15

16 Thank you for your attention! Katrin Seddig Energy Solution Center (EnSoC) e. V. Research Assistant Phone:

17 References Flath, C.; Ilg, J.; Gottwalt, S.; Schmeck, H.; Weinhardt, C. (2013): Improving Electric Vehicle Charging Coordination Through Area Pricing, Transportation Science. Waraich, R.A., M.D. Galus, C. Dobler, M. Balmer, G. Andersson und K.W. Axhausen (2013): Plug-in Hybrid Electric Vehicles and Smart Grid: Investigations Based on a Micro-Simulation, Transportation Research Part C, 28, S Kaschub, T.; Jochem, P. und Fichtner, W. (2013): Steigerung des Elektrizitätseigenverbrauchs von Heim-Fotovoltaikanlagen durch Elektrofahrzeuge, in uwf UmweltWirtschaftsForum, Vol. 21, Heft 3 (2013), S BMVBS [German Federal Ministry of Transport, Construction and City Development] (2012): Kraftfahrzeugverkehr in Deutschland 2010, Braunschweig, Germany. Jochem, P.; Kaschub, T. und Fichtner, W. (2013): How to integrate electric vehicles in the future energy system? in: Hülsmann, M and Fornahl, D. (Eds.): Evolutionary Paths Towards the Mobility Patterns of the Future, Springer, Heidelberg. ADACautotest (2012): Nissan Leaf, < abgerufen am Pfriem, M. und Gauterin, F. (2013): Less range as a possible solution for the market success of electric vehicles in commercial fleets, Proceedings of EVS27-Conference, Barcelona. Uhrig, M.; Leibfried, T. (2013): Ausbaubedarf von Parkhäusern zum Laden von Elektrofahrzeugen, in proceedings of ETG Kongress 2013, Berlin, Germany. Fraunhofer ISE (2014): Aktuelle Fakten zur Photovoltaik in Deutschland, Freiburg, Germany. BMWi [German Federal Ministry for Economic Affairs and Technology] (2013): Erneuerbare Energien, Accessed February 2014, energietraeger.html. Meteotest. Accessed March 2014, Jochem, P.; Babrowski, S.; Kaltenbach, N.; Fichtner, W. (2013): Electric Vehicle Market Penetration and Corresponding CO 2 Emissions: A German Case Study For 2030, Proceedings of IAEE Conference, Düsseldorf, Germany. 17

18 Global radiation [W/m²] Back up Global radiation for some weekdays of Stuttgart (Germany) area in Monday Tuesday Wednesday Thursday Friday February April July November Data source: Meteotest (2014) Global radiation was about 1090 kwh/m 2 /a in Stuttgart in 2013 generated energy by PV for the considered parking garage about 218 MWh 18

19 Back up CO 2 emissions of EV in different countries in 2011 Country Average CO 2 Emission in g CO 2 /kwh el g CO 2 /km* Austria Brazil Canada People s Rep. of China Denmark France Germany India Japan South Africa Sweden 17 3 Thailand Turkey United Kingdom United States World Data source: Jochem et al. (2013) * Calculated with an assumed EV electricity consumption of 0.2 kwh/km. 19