Figure 102 – uploaded by karen Sikes

See full PDF downloadDownload figure

Figure K - 12: ECAR demand with and without PHEVs on the peak week of the year — Figure 102 K - 12: ECAR demand with and without PHEVs on the peak week of the year

Related Figures (171)

Electric Utilities. In the ECAR case study, PHEVs are assumed to account for approximately 1.1 million vehicles on the road in the region in 2030. Using the defined driving patterns, the amount of battery charging needed at various times of the day was determined with Powertrain System Analysis Toolkit PSAT). The plug-in times and battery power levels were then modeled with Oak Ridge Competitive Electricity Dispatch (ORCED) model, resulting in weekly charging profiles for the vehicles (see figure below). The vast majority of power is needed during the evening hours, but smaller amounts are needed for the morning and dinner time charging. The two small spikes in the morning occur because of preconditioning for all cars and then the charging of 5% of the vehicles at work. The weekends have a higher peak on late Sunday night (early Monday morning) when all PHEVs are charging. The sharp peaks reflect the time when the high-voltage vehicles are charging as well as the low-voltage vehicles. This weekly chargin curves. Because it i that delay most charging to off-peak have occured, the overall impact on ECAR’s demand is not great. ces are not well established by 2030, and customers have no incentive to charge at nighttime, then PHEV charging could potentially have a negati However, if smart c s assumed th load. For example, he figure on their vehicles immediately followi could increase by 4, PHEV load in 2030 i 500 MW (fro is possible w and a smart grid is not a prerequ certainly aid in the process. harging practi at the market penetration is relatively low he following page demonstrates that if all ng work (from 5 p.m. to 6 p.m.) at 220V, t isite for this. However, increased use of s and smart ch ve effect on t hen peak sys m 114.9 GW to 119.4GW). It should be noted that management of the ith existing technology (e.g., vehicle-side controls or tim g profile is then added to the base system demands to create new load duration arging practices he grid’s peak PHEV owners chose to plug in em demand e-of-use pricing), mart grid tech nology would

os 23b4045 Wh an he three columns below show ECAR’s ge HEVs. Using the charging profile defined , or 0.55% of the total 600 TWh, is proj though there is a wide mix of base generation within ECAR, the added amount for PHEVs comes ainly from coal-fired power plants (withou cle plants. The renewable proportion of the added generation is from biomass co-fired with coal in e coal-fired plants. With 15% of the coal replaced by biomass, an increase in production from these $s increases both the coal-fired genera nerating capacity, base generation and added generation for in this study, a marginal increase in total generation of 3.30 ected to accommodate PHEVs in the ECAR region. t carbon capture and sequestration) and gas-fired combined ion and the biomass generation in the region.

Commercial Building Owners. For commercial vehicle-to-building (V2B) concepts to be worthwhile for the building owner, the net electricity savings will need to outweigh the monthly operating costs needed to support this program, including some form of compensation to participating vehicle owners whose battery packs are exposed to added stress. The vehicle owners will expect some form of compensation, either monetary rebates or non-monetary incentives (e.g., preferred parking spaces), in exchange for wear and tear on the battery. The net savings to the building will need to be sufficient to justify the capital and Ongoing operations cost for the program.

Table 1: A non-exhaustive list of announced plug-in electric vehicles slated for near-term production, as compiled by Sentech (last updated June 2010)

Table 3: Value propositions listed by lead investigator and required modeling tools or data

Figure 2: Network of data flow used in the regional case study analyses.

ECAR’s electricity generation mix is dominated by coal, which provided insight on how GHG emissions originating from PHEVs in this region compared to GHG emissions originating in a natural gas-dominated generation mix like in southern California. Blended electricity rates would account for peak versus non- peak rates throughout the region. PHEVs are forecasted to comprise just more than 1 million of the region’s private vehicle fleet in 2030.

Table 4: Basis for vehicle cost calculations for mid-size sedan in 2030 (2010 $)

Figure 5: Breakdown of component costs (left) and schematic of powertrain (right) for an ICE The basic architecture of the ICE is the least complex of the mid-size sedans analyzed in this case study, comprised of only the glider (vehicle minus the powertrain), engine, and transmission. (A motor/inverter, energy storage, and recharging plug/charger are not included in an ICE.) As shown in the cost breakdown in Figure 5, the estimated purchase cost for this ICE in 2030 is approximately $21,400 (using the cost equations in Table 4). A simple schematic of an ICE powertrain is also provided below. See Appendix D for detailed vehicle purchase cost calculations.

Figure 6: Breakdown of component costs (left) and schematic of powertrain (right) for HEV.

hile maintai han seen in modeling pur in this study. oattery pack PHEVs are di at home through the electric utility grid. To accommodate the increased dependence on electric power WV h a larger capacity ning an poses fferenti ated from HEVs primarily by their ability to charge with off-board electrical energy appropriate vehicle weight, the PHEV uses a battery pack wit 0 maintain a consistent performance level among the three ve oane ectrical socket for recharging. In this study, a pre-transmission he HEV. Similar to this study’s HEV, an engine of reduced size was used in the P hicle types si An inverter-integrated charging plug and charger are needed to connect the e , paralle powertrain architecture was used for the PHEV. As shown in the cost breakdown in Figure estimated purchase cost for this PHEV in 2030 is $26,925 (using the cost equation simple schematic of the PHEV parallel powertrain analyzed in this case study is sh s in Tab Appendix D for detailed vehicle purchase cost calculations. 7, th HEV fot mulatec nhanced hybrid e e 4). A own below. See Figure 7: Breakdown of component costs (left) and schematic of pre-transmission parallel powertrain (right) for PHEV.

Figure 8: Initial vehicle purchase cost comparison for ICEs, HEVs, and PHEVs produced in both 2010 and 203¢

Figure 9: Overall vehicle purchase cost with state sales tax and present value of money includea To account for common financing options and time-of-purchase fees, Figure 9 displays the overall vehicle purchase cost differences between each vehicle type in 2030 once state sales taxes and present dollar values are incorporated. The state sales tax rates for California and Ohio are assumed to be 8.25% and 5.5%, respectively. In 2030, these rates are estimated to rise steadily to 10% and 6%, respectively. For the purposes of this study, all vehicles are financed over five years at a 6% interest rate with no down payment.

Figure 10: Total fuel (liquid and electricity) costs over a ten year vehicle lifetime in southern California.

Figure 11: Total fuel (liquid and electricity) costs over a ten year vehicle lifetime in ECAR / Cleveland, Ohio The average annual amount of electricity used by a single PHEV-30 in ECAR/Cleveland, Ohio, was simulated in PSAT to be approximately 2,800 kilowatt hours (identical to southern California), adding an estimated $250 to the PHEV-30 yearly operating cost, or $2,500 to the lifetime operating cost. This additional cost still results in significant fuel savings during the lifetime of the vehicle compared to ICEs and HEVs (see Figure 11). The lifetine combined fuel (liquid and electric) costs for the PHEV-30 amount to roughly $6,750, which offers an operating cost savings of nearly $10,200 and $5,000 relative to ICEs and HEVs, respectively, in the 10-year period 4.1.2.2. Maintenance Costs It has been speculated that PHEVs will have lower scheduled maintenance costs relative to ICEs and HEVs for several reasons. First, PHEV engines are running for a lower percentage of the vehicle’s operating time, meaning they may have longer intervals between oil changes and air filter replacements. Second, regenerative braking on HEVs and PHEVs reduces wearing on the brakes and the need for brake replacements. These costs contribute significantly to a vehicle’s overall operating costs in its lifetime. As previously mentioned, the PHEVs analyzed in this study have a parallel hybrid powertrain. It should be noted that some series hybrid powertrains do not require transmissions, which could further reduce the cost to maintain a PHEVs (e.g., fluid changes). All-electric vehicles, such as

Figure 13: Overall vehicle operating cost comparison for ICEs, HEVs, and PHEVs in 2030 over a ten year lifetime in each case study location.

Figure 14: Total vehicle operating costs for each vehicle type once present value of money is incorporated. Total Vehicle Operating Cost over 10 Years

Table 5: Effect of PHEV peak shaving in July using SCE rates. Alternative charging scenarios were examined to potentially increase the amount of demand that can be shaved off of the peak loads. For example, if PHEV owners plugged in at 7 a.m. instead of 8 a.m., more vehicles could be fully charged, and the peak could be lowered by 80 kW compared to the previously mentioned 60 kW, as shown in Figure 16. The savings using the SCE rates doubles to $4,000 per month, though the number of PHEVs needed also doubles to around 70. A more detailed look at how potential benefits to commercial building owners were determined can be found in Appendix J.

Figure 15: Outcome of building owners squaring off each day’s peak by purchasing extra power in the morning to completely recharge the batteries.

Table 6: California 2030 generation and capacity factors from AEO 2008 and from ORCED simulations PHEV Value Proposition Study — Final Report / July 2010 The grid analysis for California covers the entire state, not just southern California, because the electric grid is operated as a whole. The CAISO creates a statewide market for electricity. For this analysis, the list of power plants owned by California utilities was determined from NEMS input file for EIA’s AEO 2008. The list of plants includes not only those plants within the borders but also plants owned by California utilities outside of the state, such as portions of the Palo Verde nuclear plant and the Intermountain Power project in Utah. California’s diversified mix from these plants shows a high percentage of generation from natural gas and renewables (Table 6).

Figure 17: Load Duration Curves for California loads before and after imports California’s marke ORCED, the total i import amounts were then applied to each hour based on the load in that hour as com average load for th at peak demand, imports would only be half of the amount of th imports at the mini e season. Rather than a constant amount of import each hour, it was assumed tha € average demand. Si ve demands. Th pared to the milarly, the mum demand were only 75% of the amount at the average demand in each season. This represents typical market behavior where market trading often peaks during the i demand periods. At peak times, most regions are trying to mee demands, most regions have a surplus of low-cost powe California’s imported power have reduced the demands that ge nerators experience. their own demands, while at minimum r. Figure 17 shows the LDCs before and after ntermediate shows a large amount of imports to meet demand. To simulate the net imports into mports were divided among the three seasons based on their relati e 4.3.1.3. PHEV Demand

Table 7: California study PHEV charging scenarios

By modeli created (Figure 18 the morni peaks refl (110V) vehicles. T ng the p ng and di ect the ti ug-in times and battery power levels, weekly charging profiles for the vehicles were ). The vast majority of power is needed overnight. Smaller amounts are needed for nner time charging. The weekends have larger demands in terms of kWh. The sharp me that the high-voltage (220V) vehicles are charging as well as the low-voltage he weekday demands have smaller versions of those peaks. They are not as visible because he graph displays the hourly average demand, and the high-voltage (220V) vehicles recharge in less than an hour. The weekly profile in Figure 18 is added to the base system demands shown in Appendix K. Because the market penetration is relatively low, the overall impact on demand is not great. The California PHEV growth is larger than the ECAR scenario, so the impact on the load shape is slightly more significant. Figure 19 shows the summer LDCs for the California study with and without the PHEVs. The additional demand is clearly visible in the lower portions of the LDC.

Figure 19: California summer load duration curves with and without PHEV added demand

Figure 20: California generating capacity, initial generation amounts, and added generation from PHEVs.

Table 8: ECAR 2030 generation and capacity factors from AEO 2009 and from ORCED simulations

Figure 21: Load Duration Curves for ECAR loads.

Figure 23: ECAR summer load duration curves with and without PHEV added demand

Figure 24: Net demand during peak week if all PHEV owners plug in immediately following afternoon work commute at 220V. If smart charging pract ices are not well established by 2030 and customers have no incentive to charge overnight, then PHEV charging could potentially have a negative effect on the grid’s peak load. For example, Figure 24 de following work, from 5 monstrates that if all PHEV owners chose to plug in their vehicles immediately p.m. - 6 p.m. on weekdays at 220V, then peak system demand could increase by 4,500 MW (from 114.9 GW to 119.4 GW). Offering incentives such as time-of-use pricing could encourage customers 0 shift to charging in the late evening and early morning hours.

Figure 25: ECAR generating capacity, initial generation amounts, and added generation from PHEVs Using the charging profile defined in Appendix K, the project team projects PHEVs driven in the region will cause a marginal increase in total generation of 3.30 TWh, or 0.55% of the total 600 TWh. CO2 production generation mix. Figure 25 sh additional generation from P ows the capacity and gen HEVs. Although there is a vari added amount for PHEVs comes mainly from coal-fire sequestra is from biomass co-fired with ion) and gas-fired increase in production from generation. combined cycle plants. T era ried mix ion for dp hese plants increases bo h the coal- ants (wi he renewab coal in the coal-fired plants. With increased by 2,300 metric tons, or a 1% increase. This is larger than the generation percentage increase because the added generation is more carbon intensive than the average he ECAR case study and the of base generation within ECAR, the hout carbon capture and € proportion of the added generation 15% of the coal replaced by biomass, ar fired generation and the biomass

Table 9: Fuel and engine assumptions for both regional case studies Table 10: Base and marginal electricity generation mix assumptions for both case studies in 2030. In this study, PHEVs are assumed to almost always charge on the margin.

Emissions and total energy used are divided into categories: feedstock, fuel, and vehicle operation. Each represent the energy used in the processes ranging from resource recovery to refining and distribution, and the end use in the vehicle. (It should be noted that emissions and energy originating from upstream vehicle manufacturing were not included in this study.) Energy is also broken down into categories by source: coal, natural gas, and petroleum. In some cases, negative numbers in CO2 and GHG emissions are generated and demonstrate the fact that some of the feedstocks reduce CQz rather than contributing to it.

Figure 26: Regional case study comparisons of CO2 emissions between ICEs, HEVs, and PHEV-30s in 2030. W2W CO, Emissions for Each Vehicle Type

Figure 27: Regional case study comparisons of CO2 emissions between ICEs, HEVs, and PHEV-30s in 203 Total W2W Energy Used for Each Vehicle Type 4.4.3. Increased Renewable Generation

Figure 28: PHEV-30 energy storage utilization during ten year (150,000 mile/ operation) Li-lon Battery Energy Utilization vs. Time*

Li-lon Battery Energy Utilization vs. Time* Figure 29: Potential battery pack conceptual design with only one half of the original reserve storage capacity to allow reduced battery capacity

Figure 30: Potential battery pack conceptual design without reserve capacity to accommodate cycling and aging losses

Table 11: Summary of parameters varied in this study's sensitivity analysis.

As previously mentioned, the majority of a PHEV’s operating costs savings originate from supplementing liquid fuel use with less expensive electricity. A variety of AERs were investigated in this study to see if adding larger battery packs, resulting in more electricity use, always translated to lower total ownership costs. Figure 31 indicates that total cost of ownership appears to decrease as the AER increases. The figure indicates that PHEV-30s and PHEV-40s in the ECAR region in 2030 actual have a lower total ownership cost compared to a comparable HEV in the same region, mainly because of ECAR’s low electricity prices. In the southern California region, PHEVs do not indicate a cost advantage compared to HEVs at any AER, given the assumptions taken in this study. However, PHEVs are still considered cost competitive with HEVs. In no cases investigated in this study do PHEVs have a higher cost of ownership than comparable ICEs in 2030, even though ICEs are the least expensive to purchase.

Figure 33 displays the W2W energy use for an ICE, HEV, and PHEVs with AERs of 10 miles to 40 miles in 2030 assuming a 70/30 split in E10 and E85 use. For comparison, today’s conventional ICE operating 100% on reformulated gasoline is shown. It can be concluded that ICEs easily use more energy per mile than any of the other vehicles in the study. This is expected, since the ICE has no ability to regenerate energy from braking activities, so its efficiency would be lower than that of comparable hybrids. This figure also shows the impact that vehicles with larger battery packs have on energy use. As battery size (and, Figure 32 provides a basic comparison of CO2 emissions between the three vehicle types operating in both regions in 2030, assuming a 70/30 split in E10 and E85 use in 2030. For comparison, today’s conventional CE operating completely on reformulated gasoline is shown. Many improvements have been made to the fuel economy of all three vehicle types between now and 2030, primarily because of a reduction in glider . These CQ2 emissions include those from production and use of the fuel and energy sources, but hey do not account for emissions from the manufacturing of the vehicle or its components. For the southern California region, which uses natural gas for most of its margin generation, the PHEVs investigated in this study have less W2W emissions than ICEs. However, in ECAR, where coal is used for well over half of its margin generation mix, PHEVs with an AER above 20 miles have a slightly higher production of W2W COz emission than comparable ICEs. The use of the HEV shows significantly lower production of CO2 when compared to any other vehicle in the study in either region. A breakdown of CO2 emissions for each of these cases, showing emissions originated from feedstock, fuel, and vehicle operation, can be found in Appendix G.

Figure 34: Imported oil displaced over lifetime of vehicle relative to an ICE, assuming a 70/30 split of E10 and E85 use in 2030

Figure 35: Increase in ECAR grid load due to PHEV fleets with varying AERs

Figure 38: Effects of vehicle glider weight on total W2W energy used

Figure 39: Effects of varying electricity cost of total lifetime ownership cost

Figure 40: Imported oil displaced over lifetime of vehicle relative to an ICE with the same altered drive cycles. A 70/30 split of E10 and E85 use is assumed in 2030 is assumed. The addition of 2.5% grades to portions of the base case drive cycles was investigated to see how fuel economy and lifetime fuel and electricity costs were. No hills, uphill or downhill, were included in the regional case studies. Overall, modeling results indicated negligible effects on fuel economy, even when the vehicle glider weight was increased and reduced. These calculations can be found in Appendix E.

Figure 41: Effects of varying fuel blends on CO2 emissions When the upstream energy factors for the production of biofuels are taken into account, the energy advantage still points to high concentration petroleum fuel in an efficient hybrid configuration, as shown in Figure 42. Based solely on W2W energy calculations, the HEV-E10 has a clear advantage. In the Well-to-Tank portion of the fuel cycle analysis, petroleum-based fuels have among the lowest energy losses while cellulosic ethanol has among the highest energy losses. Energy consumption in the Tank-to- Wheel portion of the fuel cycle analysis was also lower for petroleum-based vehicles than vehicles operating mostly on ethanol. However, it should be noted that mostly renewable energy is consumed in vehicles operating primarily on cellulosic ethanol. A breakdown of energy used for each of these cases, showing whether energy was used in during feedstock preparation, fuel, or vehicle operation, can be found in Appendix G.

Figure 42: Effects of varying fuel blends on total W2W energy used

Figure 43: Imported oil displaced over lifetime of vehicle relative to an ICE operating on E1¢

The effect of carbon tax may have a counter-intuitive effect on W2W CO emissions when vehicles are charged off peak, as is shown in Figure 45. The generation mix in the ECAR region generally has a high percentage of coal-generated electricity. In the case of a high carbon tax rate ($191/ton), utilities would likely change their generation mix to shift some of the coal generation from the base mix to the margin. Since the PHEVs in this study are presumed to be charged from the margin mix, the grams CO2 per mile increases relative to the base case. It is assumed in this scenario that utilities would be able to change the operating strategy of the coal-fired power generation plants to meet the margin operating strategy.

Figure 45: Effects of carbon tax rates on ECAR COz emissions

Perhaps one of the most interesting portions of this sensitivity study is the impact power generation mixes selected by regional utilities can have on W2W CO> emissi ons. More states are adopting standards that require the increased use of low-carbon or carbon-free energy sources in their mix to help reduce the carbon footprint and improve local air quality. Such methods include more utilization of nuclear power and non-carbon renewable energy resources (e.g., hydro, geot hermal, wind, solar). W2W CO2 emissions can be further reduced when coupled with a low-carbon fuel, such as E85, used to power a region’s plug-in electric vehicles. Figure 47 compares all nuclear energy, plus E85 to the projected base for both southern California all renewable energy, and all renewable energy and ECAR generation mixes in 2030. The coal-heavy ECAR region could make a change in the power generation mix to more environmentally friendly energy sources, as shown in the figure, which would reduce W2W COz emissions for transportation electricity by more than 50%. When combined with an average E85 blend, both the environmental and petroleum displacement portions of the transportation energy equation are satisfied, leaving only the cost and market availability as variables. A breakdown of CO2 emissions for each of these cases, showing whether emissions originated from feedstock, fuel, or vehicle operation, can be found in Appendix G.

The impacts of capturing and sequestering the coal on each region’s total W2W use were investigated in this study. To equip and operate a coal plant with CCS, a considerable amount of addition fuel and energy is required. Because GREET does not currently have an option for modeling CCS, estimates from the Intergovernmental Panel on Climate Change (IPCC 2005) are referenced. The panel estimates that a new supercritical, pulverized coal that uses current technology would need an additional 24%-40% of energy. Energy use associated with mining and extraction of coal would also rise. If the percentage increase was added to the ECAR region's project generation mix in 2030, total energy used in the ECAR region could rise to 4,800Btu-5,400 Btu per mile.

Figure 49: Impacts of varying market penetration rates of PHEV-30s on ECAR’s grid load

Effects of Varying Liquid Fuel Prices on Total Lifetime Ownership Cost

Figure 51: Effects of varying electricity costs on lifetime ownership cost

Public policymakers benefit from the ability to meet balance of trade, GHG emissions, and national securit goals. The most pertinent risk factor for the public sector was the potential loss of opportunities for vehicle and battery manufacturing to foreign manufacturers. Figure 53: Summary map of OEMs’ perceived risks and benefits associated with the emerging plug-in electric vehicle market

Figure 54: Summary of total cost of ownership for each vehicle type, once present value of money is applied. regions. Carbon taxes would account for a small percentage of a vehicle’s operating costs, so ebbs and flows in those rates typically do not have a strong impact. Finally, the projected cost of PHEV components (e.g., energy storage) accounts for a significant portion of the purchase price, so shortfalls in long-term cost goals may result in less financially appealing PHEVs — especially in those with high AERs and larger battery packs.

Figure 55: Net demand during peak week if all PHEV owners plug in immediately following afternoon work commute at 220V. Utilities. From a utility's perspective, the relatively slow market penetration of PHEVs assumed in this stu in combi peak de nation with smart c practices are not well estab peak de that if al 220V, th this, measures should be in place, such as time clocks with pricing or other incentives, to encourage customers to shift mands could poten en peak system de harging that shifts demand to off lished by 2030, and customers h ~peak times leads to very little impact on over mands, while providing the utility with additional sales during off-peak times. If smart charging ave no incentive to charge at nighttime, then ially have a negative effect on the grid. For example, Figure 55 demonstrates PHEV owners chose to plug in their vehicles immediately following work, from 5 p.m.-6 p.m., at mand could increase by 4,500 MW (from 114.9 GW to 119.4 GW). To avoid automatic off-peak scheduling, time-of-use 0 nighttime charging.

Table A - 1: A non-exhaustive list of announced plug-in electric vehicles slated for near-term production, as compiled by Sentech (last updated June 2010)

Table B - 1: 2030 Vehicle Parameter Assumptions

Figure C - 2: Thirteen NERC Regions, as of 2003, used in NEMS and other models. [Note: The ECAR region is no longer in operation.] Ultimately, the NERC Region formerly known as ECAR was chosen for th study for several reasons. First, ECAR inc ch were al Electric Power, FirstEnergy, and DTE Energy, with tran ISO. Many o ogy, and/or V2G technology. Thi West Virginia, whi Energy, American nterconn he benefi mix is domi region co ection and Midwes California). he effec Model's “ sufficient S of low amb highly cons , smart grid techno ich provided HG emissions origi , the colder winters in t ient temperatures on battery perform high technology” case, PHEVs are also forecasted to comprise just over 1 mill private vehicle fleet in 2030, which will supply a high enough additional load on the elec y examine the effects of PHEVs during specified charging periods. udes all or a por idered states. Sec f these organizati nsight on how G nating in a natura é site of ion of Ohio, Michigan ond, major utilities in smission ons have expressed in rd, ECA HG emissions originatin gas-dominated genera he Great Lakes area also presented the oppor ance. Finally, according to th he second regional case ndiana, Kentucky, and he region include Duke markets served by PJM erest in understanding R’s electricity generation g from PHEVs in this ion mix (southern unity to incorporate e ORNL’s MA°T he region’s id to ion of ric gr

Table D - 2: Projected purchase prices for PHEVs in 2030 when Chapter 5's sensitivity parameters for AER are applied, which range from 10 miles to 40 miles.

Table D - 3: Projected purchase prices for PHEVs in 2030 when Chapter 5’s sensitivity parameters for battery cost are applied, which range from $100 / kWh to $400 / kWh.

Table E - 1: Characteristics of each individual EPA drive cycles incorporated into this study’s work, errand and weekend/vacation trips. Table E - 2: Description of the work, errand, and weekend/vacation trips exercised by each vehicle type in this study

PHEV Value Proposition Study — Final Report / July 2010

Table E - 6: Liquid fuel and electricity cost calculations based on consumption rates — 10-, 20- and 40-mile AER instead of base case’s 30 mile AER

Table E - 12: Liquid fuel and electricity cost calculations based on consumption rates — No glider weight reduction as opposed to base case’s 30% glider weight reduction

Table E - 14: Liquid fuel and electricity cost calculations based on consumption rates - 45% glider weight reduction as opposed to base case’s 30% glider weight reduction

Table E - 16: Liquid fuel and electricity cost calculations based on consumption rates - Modified Drive Cycles Compared to Base Case

ble E - 17: Liquid fuel and electricity (if applicable) consumption calculations for ICEs, HEVs, and PHEV-30s used in each regional case study - 2.5% grades (uphill and ywnhill) as opposed to base case’s 0% grade. In addition, three different glider weights were investigated. PHEV Value Proposition Study — Final Report / July 2010

Table G - 1: Emission and energy outputs for each regional case study, modeled in GREET Key emission and energy data for each regional case study, as modeled in GREET, is compiled in Table G-1. To recap, mid-size sedan models of ICEs, HEVs, and PHEV-30s were investigated. A 70/30 split of E10 and E85 was assumed in this study; hence an average fuel blend of E30 was used for modeling purposes. The vehicles were all assumed to have a glider weight of 30% less than today’s vehicles. Furthermore, a carbon tax of $65 per metric ton has been incorporated. Finally, the generation mixes used in each case study can be found in Section 4.4.2.

Table G - 2: Emission and energy outputs for all sensitivity cases, modeled in GREET

Table H - 1: EPA Drive Cycles used in the presented study In order to simulate statistically meaningful scenarios, typical days of driving were identified, reflecting the common driving habits, and average commute time of a typical user over a year. For a complete analysis, it is also important to consider different charging availability, i.e. how often it is possible to recharge the battery. Tables H-1 and H-2 summarize the driving/charging scenarios used in this study.

Table H - 2: Driving/charging sessions during typical week

Figure H - 1: PHEV Battery and Life Estimation Model.

Figure H - 3: Battery operating regime for PHEV applications - Sizing Example Li-lon Battery Energy Utilization vs. Time*

Figure H - 4: Base Case (PHEV-30) - SOC vs. C-Rate — Trip Work A

Figure H - 5: Base Case (PHEV-30) - SOC vs. C-Rate - Trip Work B

Figure H - 6: Base Case (PHEV-30) - SOC vs. C-Rate — Trip Errand A

Figure H - 7: Base Case (PHEV-30) - SOC vs. C-Rate - Trip Errand B

Figure H - 8: Base Case (PHEV-30) - SOC vs. C-Rate -— Trip Errand C

Figure H - 9: Base Case (PHEV-30) - SOC vs. C-Rate — Trip Weekend A

Figure H - 10: Base Case (PHEV-30) - SOC vs. C-Rate - Trip Weekend B

Figures H-11 through H-13 show the effects of varying AERs on SOC vs. C-Rate operating conditions, including a PHEV-10, PHEV-20, and PHEV-40. (Only the Work A plot is provided for each vehicle.) It is clear that for smaller systems, such as the one used in the PHEV-10, the battery will be operated more often at higher damaging conditions, higher c-rate, and charge sustaining at low SOC so additional batter capacity was required for these vehicles in order to reach the required 10-year lifetime.

Figure H - 12; PHEV 20- SOC vs. C-Rate - Trip Work A Figure H - 11: PHEV 10- SOC vs. C-Rate - Trip Work A

Figure H - 13: PHEV 40 - SOC vs. C-Rate - Trip Work A

Figure H - 14: Vehicle with 0% Weight Reduction - SOC vs. C-Rate - Trip Work A Figures H-14 and H-15 show the effects of varying vehicle weight on SOC vs. C-Rate operating conditions. In this study’s base case, vehicles in 2030 were assumed to have undergone a 30% reduction in glider weight relative to today’s vehicles. The figures investigate the same vehicles with no weight reduction and a 45% weight reduction. (Only the Work A plot is provided for each vehicle.) The lighter vehicle results in less strain on the battery, resulting in an increased projected lifetime.

Figure H - 15: Vehicle with 45% Weight Reduction - SOC vs. C-Rate - Trip Work A

Figure H - 16: Extended Work Cycle - SOC vs. C-Rate — Trip Work A Moaitications to Weekly Drive Uycies Mod was mod com ified driving habits also affect SOC vs. C-Rate operating conditions. In this case, the commute to worl extended by adding an HWFET cycle between the Work A’s existing UDDS and US06 cycles. The ified Work A plot is provided in the figure. As expected, the increased VMT from a longer daily mute decreases the projected life of the battery. A scenario with no weekend cycles was also conducted, which reduced lifetime VMT, resulting in increased the battery life expectancy. (Since this scen ario did not affect the work commute, no plot is provided.

Table H - 3: Summary of battery life estimation results

* Calculations are applicable to both case studies.

Figure J - 1: Large office building (>30,000 square feet) total load in SCE. Summer Load Curve for All Large Office Buildings in SCE Territory

Figure J - 3: Fall loads for single large office building in SCE.

Table J - 1: SCE TOU-8 rates. SCE commercial rate structures are broken into both a demand and energy portion. The rates can vary based on the time of day and season in which they occur. Table J-1 shows the rates used for this analysis based on the Schedule TOU-8, Time of Use — General Service — Large published on the SCE website (SCE 2010). The summer season is June through September, while winter season is all other months. The analysis did not cover all of the intricacies of the rates, such as the combination of utility retained generating and Department of Water Resources energy rates.

Table J - 2: LADWP large general service rates.

Figure J - 4: Change in load shape for July 24-29 with PHEV charging used for peak shaving. Large Office Building Loads w/ and w/o Shifts from PHEVs When vehicles are on site, the building owner can charge them during the morning hours and thereby raise the power level for the building. In the afternoon, the building owner could drain the batteries by an equal amount, in order to lower the power level for the building. The algorithm used solved for the amount of charging needed so that the total energy was the same but the load curve was flattened across the hours from 8 AM to when the unadjusted load profile dropped below this average amount (Figure J-4).

Table J - 3: Effect of PHEV peak shaving in July using SCE rates.

Figure J - 5: Change in load shape with PHEVs plugging in beginning at 7 a.m. Large Office Building Loads w/ and w/o Shifts from PHEVs A possible conclusion from this analysis is that the savings potential, while real, is not very significant for a arge facility. $2,000 per month in savings is only approximately 1% of the building’s electricity bill, and this amount does not cover the costs of installation or operation of the charging stations. Furthermore, it does not include any payment as incentive to the PHEV owners for their likely loss of battery life. The savings per each of the 35 PHEVs works out to between $30 and $60 per month. If the building owner chose to sp! he savings with the vehicle owners, each owner would likely receive around $250 per year each. This may or may not be a sufficient amount to entice some vehicle owners to allow use of their batteries throughout he day. Encouraging earlier arrivals so charging can begin sooner will help the overall savings, but the number of required vehicles will have to increase accordingly.

Table K - 1: ECAR 2030 Generating Capacity for Three Supply Scenarios The EIA NEMS model calculates the power production and sales for the entire ECAR region as a whole through 2030. We determined the list of power plants operated within the region from NEMS input file for the Annual Energy Outlook 2009 (which are the same plants as for the HR2454 analyses). We then added the “unplanned” capacity that the model calculated as needed in each scenario from the EIA HR2454 analysis: Base, No CO2 Cost, and High CO2 Cost. Retirements were also accounted for. The resulting total capacity by technology in ORCED roughly matched the capacity defined in the each NEMS scenario (Table K-1).

Figure K - 1: Thirteen NERC Regions used in NEMS and other models

Table K - 2: Projected California Generating Capacity 2030.

Figure K - 2: AEO2009 Reference Scenario ECAR utility fuel costs The NEMS scenarios also project fuel prices for each region through 2030. Figure K-2 below shows the prices per mmBtu for each major fuel in the ECAR region. Natural gas rises to $10/mmBtu by 2030. Residual and distillate fuel oil prices, while high, were of miniscule importance for electricity generation. The prices shown include the effect of carbon cost adders. Without the carbon charge, the prices drop, especially for coal. Table K-3 shows the prices before and after carbon costs for each of the three ecanarios.

Table K - 3: ECAR 2030 Fuel Prices for three supply scenarios with and without CO2 cost.

Figure K - 3: AEO2008 Reference Scenario California utility fuel costs For the California study, the AEO2008 reference case has what some would think of as relatively low future fuel prices. Figure K-3 below shows the prices per mmBtu for each major fuel in the California region. Natural gas stays between $6 and $8/mmBtu through 2030, although current prices (and the most recent forecast in the Short-Term Energy Outlook from EIA) are $11/mmBtu. Our study doubled the AEO2008 fuel prices for 2030 as the reference prices, but sensitivities were done using the AEO2008 prices and a quadrupling of those prices.

Table K - 4: ECAR and California 2030 Base Fuel Prices with and without CO2 cost

For ECAR, the 1940 power plant units from the NEMS data sets were aggregated by technology, fuel, and variable cost, into 195 plant groups (Figure K-4). While the unscrubbed coal plants are shown, they were retired and not used in the dispatching. The hydroelectric and pumped storage capacity are not shown, but totaled 1,900 MW for hydro and 3600 MW for pumped storage. As can be seen by the blue diamonds in the figure, the lowest variable cost sources were the nuclear, renewable, and coal with carbon capture and sequestration (CCS). Variable costs include fuel, operations and maintenance, and CO2 emissions costs. (NOx and SQ2 emission costs were zero because the national emissions had fallen below the ceiling set by EPA.) The must-run plants are a variety of distributed generation and cogeneration facilities that are not directly dispatched by the utilities, so their variable cost does not enter into the dispatch decisions. Coal plants and gas combined cycle (CC) plants had similar variable costs and so were intermixed during dispatching. Gas combustion turbines (CT) were most expensive due to the price of gas (including carbon costs) and their relatively low efficiencies.

Table K - 5: ECAR 2030 generation and capacity factors from AEO2009 and ORCED

Generation capacity factors are a function of each plant group’s relative variable cost, forced outage rate, anned outage rate, and the overall dispatch of the system's plants to the system’s demands. After eveloping the reference supply and demand amounts, and dispatching them in the model, the resulting eneration, capacity factors, and percent of total were compared to the AEOQ2009 reference scenario Table K-5). Coal generation is down slightly in the ORCED run, replaced by natural gas and renewables hydro and biomass). The renewable resources capacity factor is greater than 100% because it includes iomass that is co-fired with coal in coal capacity. Ssqe@ aT

Table K - 6: California 2030 generation and capacity factors from AEO2008 and ORCED Base Demand

Figure K - 6: ECAR hourly demands for 2030 based on escalating 2006 loads.

NEMS allows imports and exports between the various regions of the country based on available transmission capacity and internal calculations on relative cost of power. ORCED does not model the other regions in order to determine economic transfers in or out. They would be a function of the relative demands and supply in each region at any point in time. Instead ORCED increases the base demand amounts to represent the exports or lowers them to represent the imports. The remainder represents the loads that are met by the plants within the region. While California shows heavy reliance on imports in the NEMS analyses, the ECAR region roughly balances its imports and exports. For 2030, there is an import amount of roughly 2% of total generation. In earlier years, there are net exports of similar amounts.

Figure K - 9: Load Duration Curves for California loads before and after imports

Table K - 7: California study PHEV charging scenarios We set that 5% of the low-V vehicles and 10% of the high-V vehicles plug in between 8am and 9am after they reach work, with a refill required of 4.6 kWh. We further set that 10% of the vehicles would plug in for an hour during dinnertime. The low-V vehicles would only fill up 1.3 kWh while the high-V vehicles could recharge their full 4.6 kWh that was used driving home. At night all vehicles plug in for charging: the 90% of them that did not charge at dinner requiring 5.3 kWh, while the 10% that did charge at dinner needing less (low-V taking 5.1 kWh, high-V just 1.8 kWh.) Over the weekend the plugging in would only be at nighttime and, according to the drive cycles used, would need 7.9 kWh to fully recharge the battery.

For this phase, we updated the charging profiles for The amount of discharge for the various trips was re-evaluated in PSAT and th following the trips was calculated (Table K-8). The s begin and end the initial plugging in of their vehicle, not the time when the population of ve examp fourth at 0615, o percen e, if the start hicles is evenly distributed over art a he PHEV beyond the Cali nd end times represe he time period in 15-minu and end times are 0600-0700 th age of vehic es that plug in for that segment, eno ne fourth at 0630 and one fourth at 0645 as $ ne fourth of the vehic fornia study in several ways. € consequent recharging nt the times when owners hey unplug it. For our calculations e increments. So, for es plug in at 0600, one . These fractions are hown in Table K-8. further downscaled by the

Table K - 8: ECAR PHEV charging scenarios On Saturday all vehicles precondition in the morning. Ninety five percent of the vehicles recharge only on Sunday evening while 5% recharge on bot Saturday and Sunday evening.

Table K - 9: PHEV charging parameters By modeling the plug-in times and battery power levels, we get a weekly charging profile for the vehicles that look like Figure K-11. The vast majority of power is needed during the nighttime. Smaller amounts are needed for the morning and dinnertime charging. The two small spikes in the morning are the preconditioning for all cars and then the charging of 5% of the cars at work. The weekends have a larger peak Sunday night when all are charging. The sharp peaks reflect the time that the high-voltage vehicles are charging as well as the low-voltage vehicles. Because the ECAR study assumes 90% of vehicles

Figure K - 13: California demand with and without PHEVs on the peak week of the year The peak demand in California occurred about a week earlier than in ECAR, using the 2006 data as a iemplate. Figure K-13 shows the addition of the PHEVs on the hourly demand during this peak week. Note hat because in the California scenario we assumed 90% of the vehicles would charge at Level-1 (110V) lhe demand is spread more evenly over the nighttime load.

Figure K - 14: ECAR summer load duration curves with and without PHEV-added demand

Figure K - 15: California summer load duration curves with and without PHEV-added demand Dispatch Results

First, a reference case was run to simulate the conditions from the HR2454cap scenario without PHEVs and the resulting wholesale electricity prices were found. These can be back-calculated to the corresponding demands to determine hourly wholesale electricity prices (Figure K-16). Prices average around 7.4 ¢/kWh, the marginal cost of many of the coal and natural gas combined cycle plants seen in Figure K-3. Daily peak prices are higher, especially in the summer when the annual peak is reached and prices climb up to 98 ¢/kWh. Because of the carbon charge of $65 per metric ton of COz, the variable costs from coal and natural gas combined cycle are around the same. As a result, prices do not stray far and depend on the price of gas and efficiency of the plant. Combustion turbines are called for only at peak times.

Figure K - 19: ECAR base capacity, base generation, and added generation with a $191/metric ton carbon cost mix of capacity and generation by 2030, even for the no-PHEV scenario. Figures K-19 and K-20 below show the base level capacity, generation, and added generation for each set of scenarios.

Figure K - 20: ECAR base capacity, base generation, and added generation with a $0/metric ton carbon cos

Figure K - 21: California generating capacity, initial generation amounts, and added generation from PHEVs using the chargin profile in Figure K-10. Figure K-21 shows the capacity and generation for base case and the added generation from PHEVs. Although there is a wide mix of generation within California, the added amount for PHEVs comes almost exclusively from gas-fired combined cycle plants. This means that PHEVs operating in California are largely being fueled by clean, efficient power plants.

Figure 102 – uploaded by karen Sikes

Related Figures (171)

Connect with 287M+ leading minds in your field