Maximilian Kloess Vienna University of Technology Gusshausstraße 25-29/373-2 Tel.: Fax.:

Transcription

1

2 THE ROAD TOWARDS ELECTRIC MOBILITY AN ENERGY ECONOMIC VIEW ON HYBRID- AND ELECTRIC VEHICLE TECHNOLOGIES AND THE INFLUENCE OF POLICIES ON THEIR DIFFUSION Maximilian KLOESS, Vienna University f Technlgy, +43 (0) , kless@eeg.tuwien.ac.at Reinhard HAAS, Vienna University f Technlgy, +43 (0) , haas@eeg.tuwien.ac.at Maximilian Kless Vienna University f Technlgy Gusshausstraße 25-29/373-2 Tel.: Fax.: kless@eeg.tuwien.ac.at 1. Overview Lw efficiency f vehicles, dependence n fssil fuels and high emissins are serius prblems f passenger vehicles tday [1]. Electrificatin f the pwertrain is cnsidered as ne f the mst prmising appraches t alleviate these prblems. Nwadays sme car manufacturers already hld hybrid vehicles in there prtfli and several thers are expected t fllw within the next years. Sme car makers even annunce the intrductin f pure electric vehicles. The cre bjective if this paper is t analyze the fllwing questins fr the example f Austria: Hw des the intrductin f hybrid electric and pure electric vehicles affect the energy system? Hw lng des it take t shift the vehicle fleet frm cnventinal drives t partly and fully electrified drives? What are the main factrs that influence this develpment? What rle can plicy play in this shift? T identify the main drivers f the diffusin f electrified vehicle prpulsin technlgies a mdel based analysis was perfrmed cnsidering six vehicles pwertrain technlgies with different degrees f electrificatin. They extend frm cnventinal internal cmbustin engines ver different hybrid electric cnfiguratins (mild hybrids, full hybrids, plug-in-hybrids, serial hybrids) t pure electric vehicles. The paper is based n a research prject supprted by the Austrian Ministry f Innvatin, Transprt and Technlgy, with the bjective t identify ptentials f vehicle pwertrain systems frm a technical, eclgical and ecnmic perspective Methds The apprach cnsists f three main steps: Firstly a detailed technical, eclgical and ecnmical analysis f the vehicle pwertrain technlgies was perfrmed t determine fuel cnsumptin, greenhuse gas emissins and investment csts [2]. In the secnd step the verall csts per kilmetre f the vehicles were analysed cnsidering fuel csts and plitical framewrk cnditins in a dynamic cntext. Thirdly the market shares and the stck penetratin f thse vehicle prpulsin technlgies were estimated based n a mdel f the Austrian vehicle stck [3]. This paper will fcus n the mdelling and the scenaris develped thrugh the mdel. Fr the mdelling a bttm-up mdel f the Austrian vehicle stck was used. Within the mdel the vehicle stck was divided in three vehicle classes, (cmpact vehicles, middle class vehicles, upper class vehicles/suvs) each class having specific characteristics, user grups and different sets f vehicle technlgies available. In the next step scenaris were develped t display the effects in medium t lng term ( ).Within this cntext the electricity cnsumptin driven by vehicle electrificatin was a matter f particular interest. 1 Supprt prgramm: A3plus - Alternative Prpulsin Systems and Fuels Prject Name: Develpment f Sceanris f the disseminatin f vehicles with partly and fully electrified pwertrain under different plitical framewrk cnditins ELEKTRA

3 Electricity (renewable) H2 (renewable) SNG Bimethane BTL Ethanl 2 Bidiesel Ethanl 1 Electricity (mix) H2 (natural gas) GTL CNG Diesel Gasline Electricity (renewable) H2 (renewable) SNG Bimethane BTL Ethanl 2 Bidiesel Ethanl 1 Electricity (mix) H2 (natural gas) GTL CNG Diesel Gasline 2.1 Mdelling Scheme The basic methdical scheme f the mdel is shwed in Figure 1. In a first step the verall csts per kilmetre f each pwertrain technlgy within each vehicle class were calculated. The relevant csts fr this calculatin were the investment csts, the fuel csts and the csts that arise frm plitical framewrk cnditins (e.g. taxes). Thse csts ad up t the verall csts per kilmetre and represent the basis f the custmer decisin. Fr the custmer decisin it was assumed that there are different levels f willingness-t-pay. Even thugh the main criteria fr the diffusin f pwertrain technlgies are the csts there are als sme restrictins that can influence custmer decisin independently frm csts. Such mandatry restrictins are fr example standards and qutas and have als been cnsidered in the mdel. In the next step the annual shares f technlgies within the yearly car registratins and the verall amunt f vehicles registered were used fr the mdelling f the verall vehicle stck. T mdel th e Austrian vehicle stck statistical data frm Austrian were used (e.g. age f the fleet, average life expectance ) T represent the ageing f the vehicle stck it was assumed that vehicle life expectancy was determined by a Weibull distributin functin. Energy Cnsumptin & Fuel mix: Bifuel Blending Fuel csts /km: Vehicle efficiency Fuel price Plitical Interventins (Standards, Qutas,..) GWh Vehicle stck (by technlgy): Investment csts /km: Acquisitin csts Vehicle life time Interes level Kilmters driven by year Overall csts /km: Purchase decisin Annual vehicle registratin By pwertrain system Overall vehicle stck: Tax n nwnership Tax n acquisitin Fuel tax subsidies vehicle stck Fuel cell Elctric drive H2-Mtr-Hybrid CNG-Hybrid Diesel-Hybrid Gasline-Hybrid H2-Mtr CNG-Mtr Diesel-Mtr Gasline-Mtr Taxes /km: Tax n nwnership Tax n acquisitin Fuel tax subsidies Diffusin Barriers t CO2-equivalent Greenhuse gas emissin (WTW, TTW, WTT: Figure 1: Scheme f the mdelling T mdel the yearly demand fr vehicles and therefre the verall car registratins an energy ecnmic apprach was applied: It was assumed that the develpment f annual vehicles registratins depends n the level f incme, expressed by the GDP, the fuel price level and the level f investment cst. Thus the develpment f the annual vehicle demand is given by the elasticity f fuels price ε FP, elasticity f vehicle price ε IC and incme elasticity εy. These elasticities have been derived frm ecnmetric time series analyses. Z ε FP ε IC Z FP IC t + 1 ε t + 1 ε IC t+ 1 = t ( ) ( ) FPt ICt = 0,5 = 1 GDPt ( GDP FP + t 1 ) ε γ

4 ε γ = 0,8 Z.vehicle registratin per year IC investment csts FP fssil fuel price level GDP incme level represented by the grand dmestic prduct ε FP price elasticity ε IC elasticity n investment csts ε y.incme elasticity 2.2 Reference Vehicles As mentined abve three classes f vehicles were distinguished: Cmpact class vehicles Middle class vehicles Upper class vehicles (including SUVs) Each vehicle class was defined by vehicle mass and engine pwer. T determine thse specificatins and the allcatin f each class within the vehicle stck and the yearly vehicle registratins fr the particular case f Austria histric data was used. Within the vehicle classes different pwertrain technlgies were cnsidered including the entire spectrum frm cnventinal drive with internal cmbustin engine t pure electric drive. The fllwing systems were analysed: Cnventinal Drive Micr Hybrid Mild Hybrid Full Hybrid Plug-In-Hybrid Serial Hybrid (with plug-in) Electric Vehicle Fuel Cell Vehicle The specific characteristics f the vehicles are shwed in Table 1. Cnventinal Drive Gasline Cnventinal Drive Diesel Mild Hybrid Full Hybrid Plug In Hybrid Serial Hybrid EV FCV Vehicel Weight (kg) Cmbustin Engine Pwer (kw): Electric Mtr Pwer (kw): Li-Inen Batteries (kwh) Electric Range (km) Overall Range Table 1: Vehicle Pwertrain Specificatins (middle class) 2.3 Fuel csts Fr the fuel csts f the vehicles tw parameters are imprtant: The fuel efficiency f the prpulsin system and the fuel prices.

5 2.3.1 Fuel efficiency: The data fr the fuel efficiency f fr all vehicles with different prpulsin systems was prvided by ur prject partner AVL 2 wh is an expert n the field f autmtive Research and Develpment [2]. In the fllwing chart the tank-t-wheel (TTW) efficiencies f different pwertrain systems are shwed. It is bvius that the efficiency increases with the degree f electrificatin f the prpulsin system. 90% 80% Efficiency η 70% 60% 50% % 30% 20% 10% 0% Figure 2: TTW Vehicles Efficiency 2010 and Fuel prices: It was assumed that there wuld be a cntinuus increase in the price f vehicle fuels in the time frame due t the increase f fssil fuel prices. Fr the analysis tw basic fssil fuel price scenaris were determined bth starting at the 2009 price level in 2010: Lw-Price-Scenari: in this scenari it is assumed that the fssil fuel price increases in a mderate manner. High-Price-Scenari: in this scenari it is assumed that there wuld be a strng increase f the fssil fuel price level up t AVL LIST GMBH

6 0,200 0,180 /kwh 0,160 0,140 0,120 0,100 0,080 Gasline lw Diesel lw CNG lw Electricity mix lw Gasline high Diesel high CNG high Electricity mix high 0,060 0,040 Figure 3: Net Fuel Price Scenaris 2.4 Investment csts The analysis f the investment csts is based n the csts f the single cmpnents. The csts f the cmpnents were subject f detailed analysis within the prject [2]. Figure 4 shws hw the different cmpnent grups cntribute t the verall cst f the each vehicle technlgy. It shws that the verall investment csts increases with the degree f electrificatin f the prpulsin system. The main driver f the investment csts are the batteries that are required t permit the desired electric range Basic Vehicle IC Engine+Transmissin Electric Drive Batteries FC + H2 Tank Figure 4: Net csts f middle class vehicles 2010 Therefre the csts f batteries (Lithium In Batteries) can be seen as the crucial factr within the investment csts f electrified vehicles. Their specific csts in 2010 were assumed t be at 700 /kwh. Fr the mid- t lng term develpment f the csts f batteries the cncept f technlgical learning

7 was applied. T estimate the cumulative prductin f lithium in tractin batteries the scheduled intrductin f majr autmtive prducers was used fr the time Fr the years it was assumed that the lithium in batteries becme the standard technlgy fr bth electric and hybrid electric vehicles and therefre fllws the characteristic S-shape f a technlgical substitutin prcess. Furthermre it is assumed that the trend twards hybridisatin f vehicles was a rbust prcess and wuld cntinue within the fllwing decades. The mathematical descriptin f the learning curve is given by the expnential functin. C( x) = a * x b C...cst per unit [EUR/unit] a...cst f first unit prduced [EUR/unit] x...number f prduced units b...learning index [-] It is evident that the cst reductin depends n the glbal cumulative prductin and the learning index. Frm the learning index the prgress rati p can be calculated. The prgress rati expresses what cst reductin wuld be caused by a dubling f cumulative prductin (percentage f frmer csts after a dubling f prductin). p = 2 b p...prgress rati [-] Fr the case f lithium In batteries a prgress rati f 92,5% was assumed. The cst reductin f lithium In batteries frm 2010 t 2050 is depicted in figure Figure 5. /kwh Csts f Li in Batteries Figure 5: Csts f Li-In Batteries The reductin f battery csts effect a significant reductin f verall vehicle csts f hybrid and electric vehicles up t 2030 and 2050 (see Figure 6 & Figure 7). Therefre technlgical develpment f batteries has significant influence n their medium t lng term diffusin.

8 Basic Vehicle IC Engine+Transmissin Electric Drive Batteries FC + H2 Tank Figure 6: Net csts f middle class vehicles Basic Vehicle IC Engine+Transmissin Electric Drive Batteries FC + H2 Tank Figure 7: Net csts f middle class vehicles Plitical framewrk cnditins Plicy can play an imprtant rle fr the diffusin f efficient vehicle pwertrain systems in the passenger car sectr and therefre fr the energy efficiency and the greenhuse gas emissins f the sectr as whle. The plitical framewrk cnditins cnsidered in this paper were thse that influence the cst f transprt directly (mainly taxes) and thse that influence the price indirectly thrugh

9 cmpulsive restrictins (e.g. standards and qutas) Fr the present analysis nly the first grup was cnsidered. The main taxes that are imprtant fr the passenger vehicle sectr in Austria are: Tax n Acquisitin Tax n Ownership Tax n Fuels Tax n Acquisitin: this tax depends n the fuel cnsumptin and the greenhuse gas emissins f the vehicle Tax n Ownership: the height f this tax depends n the engine pwer f the vehicle Tax n Fuels: in Austria fuels are taxed by unit ( /litre r /kg) It was assumed that the plitical framewrk cnditins in 2010 are the same as they are nw (2009). Fr the time frame tw main plicy scenaris were defined: Business as usual scenari BAU-Scenari: In this scenari it was assumed that the plitical framewrk cnditins remain n the 2010-level Plicy scenari: it was assumed that taxes wuld be adapted with regard t an increase in energy efficiency and a reductin f greenhuse gas emissins. Therefre taxatin f hydrcarbn fuels and inefficient vehicles gets raised stepwise within the perid. 2.6 Overall Csts The service csts f each vehicle cnsists f it s fuel csts and it s capital csts. Thse csts are added up as depicted in Figure 1. The mathematical definitin is expressed by the fllwing equatins: SC j = EC + CC [EUR/km] j SP _ j SC...service csts [EUR/km] EC....energy csts [EUR/km] CC SP_j...specific capital csts f vehicle j [EUR/km] EC = FC * FP [EUR/km] j j j FC j.energy cnsumptin f vehicles j [kwh/km] FP..fuel price [EUR/kWh] CC + SP _ j = ( α *( IC j + TAj ) *(1 + VATj )) / Skm TO [EUR/km] j α DT ( r * (1 + r) (1 + r) 1 = DT r interest rate DT...depreciatin time α... Annuity factr IC j Investment csts f vehicles j (EUR/vehicle) TA j..tax n acquisitin TO j...tax n wnership VAT j Value added tax S km.driven kilmetres by year

10 In the mdel the ecnmic perfrmance f all vehicle classes with all prpulsin systems was calculated at different yearly kilmetrage rates. Fr the further emplyment in the mdel it was the csts were calculated in per kilmetre ( /km). Fr the analysis an interest rate f 5% and a depreciatin time f 14 years were cnsidered. The fllwing figures shall give an verview n the develpment f verall csts in the time frame 2010 t Fuel Csts Investment Csts Figure 8: Yearly verall csts f vehicles at km/a Fuel Csts Investment Csts Figure 9: Yearly verall csts f vehicles at km/a 2050 (Plicy Scenari and high fuel price) Figure 8 and Figure 9 indicate that tday hybrid and electric vehicles are nt cmpetitive frm an ecnmic pint f view, neither in the middle class segment nr in the cmpact class. Their high investment cst cannt be cmpensated by the fuel savings.

11 Until 2050, due t the reductin f the investment csts and the shift within the plitical and ecnmic framewrk cnditin (Plicy Scenari + high fssil fuel prices), the cmpetitiveness f electrified vehicles imprve strngly. By that time they are catching up with cnventinal technlgies and sme becme even better. 2.7 Scenaris T analyse the effect f plicies and different fssil fuel price levels tw basic scenaris were develped. The scenaris defer frm each ther in the way f the plicies implemented and the fssil fuel price levels (see chapter and 2.5): Scenari A: BAU-Plicy & Lw fssil fuel level Scenari D: Active Plicy & High fssil fuel price level 3. Results Using the mdel it was pssible t investigate the effectiveness f plicies n the diffusin f efficient vehicle technlgies n the ne hand and the impact f further drivers, such as fuel prices and develpment f vehicle technlgies n the ther hand. The results f the mdel were als used t shw the effects f the afrementined drivers n the energy cnsumptin and the greenhuse gas emissins. 3.1 Scenari A Figure 10 shws the develpment f the vehicle stck in Scenari A. In this scenari the taxatin level remains the same in the entire perid In the same time the imprving vehicle efficiency cmpensates the light increase f fssil fuel prices. Therefre the csts f passenger stay at a lw level r even decrease causing the demand fr transprtatin t increase and the vehicle stck t grw. In the case f Austria that means that the vehicle stck grws frm 4,2mi vehicles t abut 5,8mi vehicles Cnventin Drive Micr Hyb. Mild Hybrid Full Hybrid Plug In Hybrid Serial Hybrid EV FCV Figure 10: Develpment f the Vehicles Stck Scenari A Taking a lk at the pwertrain systems within this stck it shws that cnventinal drive systems get replaced by Hybrid ptins which becme the standard technlgy in the mid t lng term. In this scenari Plug-In-Hybrids and Electric Vehicles dn t achieve big shares nt even in the lng term.

12 In Scenari A the final energy cnsumptin f the passenger sectr decreases slightly (see Figure 11). Althugh the vehicle fleet increases the higher vehicle efficiency causes a slight reductin f verall cnsumptin. Regarding the energy carriers ne can see that liquid hydrcarbns remain the mst imprtant energy carriers even in a lng term even thugh there is a cnsiderable share f cmpressed natural gas and sme electricity. 40,0 35,0 30,0 TWh 25,0 20,0 15,0 Gasline Diesel CNG Electricity 10,0 5,0 0,0 Figure 11: Final Energy Cnsumptin and Energy Carriers - Scenari A In Figure 12 the tank-t-wheel (TTW) greenhuse gas emissins f the entire vehicle fleet are depicted. It shws nly a slight decrease in emissins stabilising at a level f abut 8 000t per year t CO2 eqivalent Figure 12: Tank-t-Wheel (TTW) greenhuse gas emissins Scenari A

13 3.2 Scenari D In Scenari D taxatin f vehicles and fuels causes a rise in verall transprtatin csts which reduces the demand fr passenger transprt. This develpment causes the vehicle stck t stabilise at the level f Regarding the vehicle technlgies within the stck it shws that cnventinal drive systems get replaced by hybrid systems in a shrt- t mid term. In a lng term highly electrified vehicles such as Serial Hybrids, Plug-In Hybrids and Battery Electric Vehicles cnsiderably gain share in the verall vehicle stck. Tgether they reach almst tw thirds f the vehicles stck in 2050 (Figure 13) vehicles Cnventin Drive Micr Hyb. Mild Hybrid Full Hybrid Plug In Hybrid Serial Hybrid EV FCV Figure 13: Develpment f the Vehicles Stck Scenari D The final energy cnsumptin in scenari D is decreasing by abut 50% up t 2050 caused by tw develpments: firstly there is a general switch t mre efficient vehicles while the fleet size stays abut the same. Secndly the deplyment f electricity as final energy carriers causes a strng reductin in final energy cnsumptin in a lng term. Nevertheless it als has t be cnsidered that in this chart primary energy cnsumptin fr the prductin f the electricity is nt depicted. In an verall energy balance the reductin wuld nt be that brad (Figure 14). 40,0 35,0 30,0 TWh 25,0 20,0 15,0 Gasline Diesel CNG Electricity 10,0 5,0 0,0 Figure 14: Final Energy Cnsumptin and Energy Carriers - Scenari D In Figure 15 the TTW greenhuse gas (GHG) emissins f the sectr are depicted fr scenari D. One can see that similar t the energy cnsumptin there is als a strng decrease f GHG emissins. As

14 these are TTW figures, emissins frm electricity prductin are nt included in this chart. In the Well-t-Wheel balance the reductin wuld be highly dependent t the surce f electricity t CO2 eqivalent Figure 15: Tank-t-Wheel (TTW) greenhuse gas emissins Scenari D The Results indicate that generally the diffusin f electrified vehicles is quite rbust. Taking a clser lk it can be nticed that diffusin f vehicles with higher degrees f electrificatin (e.g. serial hybrids r pure electric vehicles) vary strngly in different scenaris. In particular plitical framewrk cnditins can have significant effects n the diffusin f electric vehicles. Due t the high investment cst f such types f vehicles they can nly cmpete under favurable framewrk cnditins especially in the shrt run. On a lng term electrificatin f the passenger vehicle fleet can cause a majr shift in energy demand f the entire sectr. On shrt t medium term liquid hydrcarbns remain the main energy carrier fr the sectr.

15 4. Cnclusins The main cnclusins f this analysis are: Electrificatin f vehicles is abut t cnsiderably gain market share within the next years in all Scenaris. Especially micr and mild hybrid cnfiguratins becme a standard technlgy in the near- t medium-term. They can be seen as a lgical step in autmtive develpment t increase verall vehicles efficiency while keeping the internal cmbustin engine as the main prpulsin cmpnent and liquid fssil fuel the standard energy carrier. Their ptential f reducing the energy demand and the greenhuse gas emissins f the vehicle fleet hwever are limited. In rder t reduce energy demand and emissins mre cnsiderably, vehicles with higher degrees f electrificatin such as Plug-In Hybrids and Electric Vehicles are required. T prmte the diffusin f thse types f vehicles in a shrt term strng plitical measures are necessary. Their lng term perfrmance will highly depend n the technlgical develpment and the csts f batteries, representing the key cmpnent f thse vehicles. This technlgical prgress f battery technlgy will be strngly affected by the shrt term perfrmance f eletrified vehicles due t technlgy learning effects. Furthermre higher taxatin f individual mtrised transprt effects a stabilisatin f the vehicle stck n the level f 2010 facilitating reductins in verall energy cnsumptin and greenhuse gas emissins thrugh the emplyment f efficient prpulsin technlgies. Regarding the ptential electricity cnsumptin f the vehicle fleet in different scenaris it turns ut that the increase in verall demand is practically negligible in the shrt term. In the medium t lng term they nly cause a slw rise f verall electricity cnsumptin ver a lng perid leaving electricity prducers enugh time t adapt their capacities. References [1] Internatinal Energy Agency IEA, Energy Technlgy Perspectives 2008 [2] M. Kless, A. Weichbld, K. Könighfer: Technical, Eclgical and Ecnmic Assessment f electrified pwertrain systems fr passenger cars in a dynamic cntext (2010 t 2050), The Internatinal Battery, EVS-24 Hybrid and Fuel Cell Electric Vehicle Sympsium, Stavanger 2009 [3] M. Kless, A. Ajanvic, R. Haas, A dynamic analysis f the ecnmics f alternative vehicles and fuels in Austria up t 2050, Third Internatinal Cnference n Eclgical Vehicles & Renewable Energies EVER Mnac, 2008