Agent-Based Modelling of Electric Mobility Adoption
•
1 recomendación•370 vistas
The document describes the ABCD (Agent-based Buying Charging Driving) model. The ABCD model is an agent-based model that simulates the future of electric mobility in the Netherlands by modeling the behaviors of individual agents such as consumers, vehicles, and charging points. The model takes a bottom-up approach to understand phenomena from the behaviors and interactions of individual agents. It is used to study topics like the adoption of electric vehicles, impact on the electricity grid, and optimal policy interventions over time.
Recomendados
Recomendados
Más contenido relacionado
La actualidad más candente
La actualidad más candente (20)
Similar a Agent-Based Modelling of Electric Mobility Adoption
Similar a Agent-Based Modelling of Electric Mobility Adoption (20)
Más de Vincent Everts
Más de Vincent Everts (20)
Último
Último (10)
Agent-Based Modelling of Electric Mobility Adoption
- 1. ABCD model Agent-based Buying Charging Driving model Auke Hoekstra Senior advisor smart mobility TU/e Strategic consultant ElaadNL, Alliander, Urgenda, FET & NKL E-mail: auke@aukehoekstra.nl Twitter: @aukehoekstra Future Smart Mobility ECOMOBIEL Auke Hoeksra, Senior Advisor Electric Mobility TU/e a.e.hoekstra@tue.nl @aukehoekstra
- 2. ABCD model Agent-based Buying Charging Driving model Auke Hoekstra Senior advisor smart mobility TU/e Strategic consultant ElaadNL, Alliander, Urgenda, FET & NKL E-mail: auke@aukehoekstra.nl Twitter: @aukehoekstra Modelling the future of Electric Mobility in the Netherlands Auke Hoeksra, Senior Advisor Electric Mobility TU/e a.e.hoekstra@tue.nl @aukehoekstra
- 5. The Future of Mobility is: Electric Self Driving Shared 512-10-2017
- 6. The Future of Mobility is: Electric Self Driving Shared 612-10-2017
- 7. Automotive Technology Sqiffer questions • When will electric cars become cheaper? • Why is the motor so much better? • Do electric cars really emit less CO2? • What is the role of hydrogen? • How can EVs boost the introduction of off-shore wind? • How can EVs boost the introduction of solar panels? • How can EVs threaten or help electricity grid stability? 712-10-2017
- 8. The Future of Mobility is: Electric Self Driving Shared 812-10-2017
- 9. / name of department I know what I’m taking about with my Nissan Leaf (120km range): range anxiety is real! 912-10-2017
- 10. 1012-10-2017
- 11. Hier is een snellader buiten gebruik 1112-10-2017
- 12. Deze snellader is kapot 1212-10-2017
- 13. Thuis net niet gehaald 1312-10-2017
- 14. Thuis helemaal niet gehaald 1412-10-2017
- 15. / name of department 1512-10-2017
- 17. 2. Increase in usage of Electric Vehicle (EV) 17
- 18. 2. Increase in usage of Electric Vehicle (EV) 18
- 19. 2. Increase in usage of Electric Vehicle (EV) 19
- 20. 2. Increase in usage of Electric Vehicle (EV) 20
- 21. 2. Increase in usage of Electric Vehicle (EV) 21
- 22. 2. Increase in usage of Electric Vehicle (EV) 22
- 23. 2. Increase in usage of Electric Vehicle (EV) 23
- 24. 2. Increase in usage of Electric Vehicle (EV) 24
- 25. 2. Increase in usage of Electric Vehicle (EV) 25
- 26. 2. Increase in usage of Electric Vehicle (EV) 26
- 27. 2. Increase in usage of Electric Vehicle (EV) 27
- 28. 2. Increase in usage of Electric Vehicle (EV) 28
- 29. 2. Increase in usage of Electric Vehicle (EV) 29
- 30. Automotive Technology Why will electric vehicles take over? (the extremely simplified version) • 1/3 of the energy use => 1/3 of energy cost 250.000 km in VW Golf = 22.000 liter = 35.000 euro Idem in EV is 10.000 euro: you save 25.000 euro. • Motor doesn’t need maintenance => 1/3 of maintenance costs You save 7.000 euro (Together with energy 32.000) • Battery for 350 km range (65kWh, consumer price): 2007: 100.000 euro (you lose 65.000) 2017: 25.000 euro (you save 10.000) 2027: 10.000 euro (you save 25.000) • Bonus: sports car motor with the price and efficiency of a regular car motor 3012-10-2017
- 31. It’s all about learning curves 3112-10-2017
- 32. Learning curves driven by feedback loops Here you see what goes wrong if you don’t use learning curves and feedback loops. And the IEA is the most influential organisation in the world in the field of energy! 3212-10-2017
- 33. Automotive Technology Battery price developments 3312-10-2017 Source: “Rapidly falling costs of battery packs for electric vehicles” by Björn Nykvist and Mans Nilsson, Nature Climate Change, March 2015
- 34. Automotive Technology We have researched this intensively 3412-10-2017 Predicted Battery Price Price€/kWh 0 300 600 900 1200 Year 2000 2008 2015 2023 2030
- 35. Automotive Technology Not only using extrapolation but also by using bottom up analysis of raw materials and processes used. 3512-10-2017 0 25 50 75 100 Battery Cost breakdown 5 4 9 4 4 4 5,0 16 49 Materials Purchased Items Direct labour Variable overhead General, sales, administration R&D Depreciation Profit Warranty Source: R. Kochhan, et. Al., “An Overview of Costs for Vehicle Components, Fuels and Greenhouse Gas Emissions,” Publ. Www Res. Net, Feb, 2014. Rest of vehicle 60,0 Battery Pack 40,0
- 36. Automotive Technology It shows that even current day batteries can become 5-10 times cheaper ($50-$100 in 2050). If solid state batteries or metal air batteries break through things will become even more interesting. 3612-10-2017 Li-S LR-NCM NCM 0 75 150 225 300 Raw material Process Overheads Source: Adapted from “Rechargeable Batteries: Grasping for the Limits of Chemistry” by Petr Novak, Journal of Electrochemical Society, October 2015
- 37. Automotive Technology Battery price developments 3712-10-2017
- 38. So how come EVs are getting cheaper? The motor is the answer. 3812-10-2017 ICE Electric 1,6 kW/kg 4,3 kW/kg 3x more powerful per kg 0,4kW/lr 13,6 kW/lr 40x smaller per kW 20-25% efficient 90-98% efficient 3x more efficient Many moving parts 1 moving part morerelialble Leuk!
- 39. Two ways to bring 300 kW to the wheels (BMW and Tesla to scale) 3912-10-2017
- 40. Two ways to bring 300 kW to the wheels (BMW and Tesla to scale) 3912-10-2017
- 41. EV: fast cars are no longer inefficient Bugatti Veyron (0-100 2.5 sec – 2 pers) in the city = 40l/100km (4kWh/km) Tesla (0-100 2,3 sec – 7 pers) in the city = 0,2 kWh/km 4012-10-2017 So a VERY rough calculation of battery size in kWh is: desired range in km / 5
- 42. Automotive Technology Drive train cost developments 4112-10-2017 We had to adjust findings to observed market cost. So again double analysis: bottom-up from parts and top down from market prices.
- 43. Automotive Technology Construct virtual cars to choose from 4212-10-2017
- 46. Automotive Technology Include residual value 4512-10-2017
- 47. Automotive Technology Discount rate • We chose a discount rate of 5% • However, even a discount rate of 10% moves the break even point by less than a year in most scenarios 4612-10-2017
- 48. Automotive Technology Energy and maintenance cost 4712-10-2017 Very roughly put: energy and maintenance are both 1/3
- 49. Automotive Technology With average km per year 4812-10-2017
- 50. Automotive Technology With more km per year 4912-10-2017
- 51. Automotive Technology Second hand market 5012-10-2017
- 52. Automotive Technology Conclusions • FEVs trump PHEVs because higher sticker price, energy costs and maintenance costs is outweighted by lower battery prices • New cars (25000 km/y) wil become profitable without subsidies before 2020 in the Netherlands • Second hand cars are even more profitable • Adoption speed will be driven by awareness of TCO, availability of charging infrastructure, and availability of cars • New types of models can more accurately predict adoption, impact on (renewable) electricity grid and optimal interventions 5112-10-2017
- 53. Automotive Technology Electric long haul trucks? (Is the Tesla semi believable?) 5212-10-2017 https://www.livinglabsmartcharging.nl/nl/Auteurs/auke-hoekstra
- 54. Heterogeneous actors by using agent-based modelling: “play out” the real life system in the computer. 5312-10-2017 ▪ Bottom-up modelling: system not defined in advance. ▪ Every agent is unique and you can have many thousands of them. ▪ De-facto standard for socio- technical systems [4]. ▪ Agents can be people but also institutions, machines or buildings. [3] https://www.anylogic.com/use-of-simulation/agent-based-modeling/ [4] K. H. Van Dam, “Capturing socio-technical systems with agent-based modelling,” PhD Thesis, TU Delft, Delft University of Technology, 2009. Source: [3]
- 55. Automotive Technology Future work: better interaction with neighborhood An example neighborhood in the ABCD model. 5412-10-2017
- 59. Bottom-up approach 5512-10-2017 Roads, exits and parking places Neighborhood (determined by NKL)
- 60. Bottom-up approach 5512-10-2017 Residents, EVs and CPs Roads, exits and parking places Neighborhood (determined by NKL)
- 61. Bottom-up approach 5512-10-2017 Residents, EVs and CPs Roads, exits and parking places Neighborhood (determined by NKL) 1. Zeeheldenkwartier region, The Hague 2. Alteveer region, Arnhem
- 65. These trends are good, but cause problems in the Dutch energy system… 57
- 66. These trends are good, but cause problems in the Dutch energy system… Mismatch! 57
- 67. Also: charging when electricity is expensive 58
- 68. Solution: shift charging load! (smart charging) 59
- 69. Smart Charging uses more renewable energy and is also cheaper 60
- 70. Average local charging load in Zeeheldenkwartier, The Hague 2017 2023 61
- 71. Average local charging load in Zeeheldenkwartier, The Hague 66 kWh 334 kWh 2017 2023 61
- 72. Charging costs in Zeeheldenkwartier, The Hague 62 2017 2023
- 73. Charging costs in Zeeheldenkwartier, The Hague 26,41 % 52,12 % 62 2017 2023
- 74. Renewable charging during the winter in Zeeheldenkwartier, The Hague 63 2017 2023
- 75. Renewable charging during the winter in Zeeheldenkwartier, The Hague 63 28% growth 28% growth 2017 2023
- 76. The Future of Mobility is: Electric Self Driving Shared 6412-10-2017
- 77. The Future of Mobility is: Electric Self Driving Shared 6512-10-2017
- 90. Automotive Technology 7012-10-2017 Over 30 jaar mag je alleen nog maar rijden op het circuit van Zandvoort.
- 93. The Future of Mobility is: Electric Self Driving Shared 7312-10-2017
- 94. The Future of Mobility is: Electric Self Driving Shared 7412-10-2017
- 95. Automotive Technology Vehicles will also become smaller because most trips are 1 person trips: cheaper, less PM, serious accidents, roads and parking 7512-10-2017
- 102. Automotive Technology Conclusions 1. Elektric vehicles will be cheaper and more fun to drive 1/3 of energy and maintenace costs & cheap battery 2. Cars will become autonomous in 30 years you can only drive yourself on Zandvoort: advantages are less accidents and roads and more time 3. Cars will be shared Spacious, green, healthy, silent & safe cities with better mobility at lower costs 7812-10-2017