Printer Friendly

U.S. Light-Duty Vehicle Air Conditioning Fuel Use and Impact of Solar/Thermal Control Technologies.

Introduction

Off-cycle emissions credits are provided as part of the "Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards for Vehicle Model Years 2017-2025" [1]. These off-cycle credits are intended to capture the efficiency improvements by technologies that are not captured in standardized vehicle drive cycle testing. Within the off-cycle credit menu, solar/thermal technologies that improve air conditioning (A/C) system performance are listed and include solar reflective glass, solar reflective paint, and both active and passive ventilation. The listed carbon dioxide (CO2) emissions credits provided for the solar/thermal technologies were determined by the U.S. Environmental Protection Agency (EPA) and National Highway Traffic Safety Administration (NHTSA) based on both the baseline fuel use associated with vehicle A/C and the expected incremental performance gains of the technology [2]. A description of the methods and assumptions used by the U.S. EPA and NHTSA to establish baseline A/C CO2 emissions and solar/thermal credits can be found in the published joint technical support document [2]. The U.S. EPA determined A/C emissions using simulations of a vehicle driven over an SC03 drive cycle using a scaled fixed displacement A/C compressor. The simulations also used an A/C system power consumption curve at 27[degrees]C and 60% relative humidity as representative of the U.S. environmental conditions during A/C operation. Although the credits have been established for regulatory purposes, independent computation of these critical values provides a valuable comparison and establishes a method for evaluating future technologies in this area.

Performance evaluation of vehicle A/C systems and the associated impact of solar/thermal technologies on a national scale are a significant challenge because their effect is dependent on local environmental conditions, vehicle usage patterns, A/C system operation, and powertrain performance. The time scales of interest for the vehicle cabin and external environment, A/C system vapor compression cycle, and vehicle powertrain vary from hours to milliseconds, making co-simulation of the systems difficult for a broad range of input conditions. In addition, subsystem model complexity and input permutations significantly impact computational requirements necessary for their evaluation.

A light-duty vehicle A/C fuel use method was previously developed by Rugh et al. [3] that incorporated rigorous environmental conditions, thermal comfort correlations, A/C system performance correlations, and a vehicle model to estimate the national A/C fuel use for light-duty vehicles. However, this method incorporated simplifications for vehicle cabin load estimations and A/C system performance that limit its value for A/C technology evaluations. Recent work by Kambly and Bradley [4] estimated the impact of heating and cooling on plug-in electric vehicles. Their work utilized a simplified vehicle cabin thermal model and fixed A/C and heating system performance for impact estimations. Their work followed a similar methodology and leveraged environmental and vehicle usage pattern data common with the approach described herein. However, their approach was limited to electric vehicle performance. Finally, the Green-MAC-LCCP calculation tool [5] was developed to calculate both direct and indirect emissions of mobile A/C refrigerants, capturing vapor compression cycle system performance for a range of conditions. However, because the tool does not include a vehicle cabin model, extension of the tool to estimate the impact of solar/thermal technologies with broadened input conditions was determined to be impractical.

Although no standardized method exists that incorporates the described elements for A/C system and solar/thermal technology performance evaluation, several vehicle system tools such as vehicle cabin thermal models and complete A/C system models are available. This article documents the development of a process that incorporates existing system models to evaluate national-level baseline A/C fuel use and the expected benefit of solar/thermal technologies.

National-Level A/C Analysis Methodology

The light-duty A/C fuel use analysis methodology calculates A/C system energy flows starting with the cabin thermal energy required to maintain a comfortable environment in the passenger cabin and ending with the fuel consumed. The methodology leveraged an existing method for heavy-duty vehicle rest-period climate control system performance evaluation [6]. A process diagram showing the major components of the developed method and associated key inputs is provided in Figure 1. The vehicle cabin model, A/C system model, and vehicle propulsion models used in the process were previously developed by the National Renewable Energy Laboratory (NREL). These tools include CoolCalc, a vehicle cabin heating, ventilating, and air conditioning (HVAC) load estimation software [7]; CoolSim, a MATLAB/Simulink thermal modeling framework [8]; and Future Automotive Systems Technology Simulator (FASTSim), a high-level vehicle powertrain analysis tool [9]. Further details on key components are provided in subsequent sections.

Vehicle Cabin Thermal Model

The CoolCalc vehicle cabin HVAC load estimation tool builds on the EnergyPlus thermal zone solver, using single-zone air nodes and simplified model geometry to perform large-scale computations while limiting computational demand. A detailed description of the tool is provided by Lustbader et al. [10]. The model simulates heat transfer between the vehicle cabin and an external environment defined by Typical Meteorological Year weather data for select locations in the United States [11]. The model provides A/C evaporator capacity requested to maintain the interior air temperature at a user-defined setpoint at 1-minute time steps for the associated annual weather location specified. SUV, compact, and mid-size vehicle size classes were selected for the analysis to represent the U.S. light-duty fleet. Selection of the classes was based on available validation data, range of cabin volumes, and powertrains. Detailed vehicle size class registrations were obtained from Polk automotive data [12] and were assigned to the three generic classes to provide weighting factors. SUVs represented 52% of the fleet, followed by mid-size vehicles at 30%, and compact vehicles at 18%. The light-truck category represented 19.8% of all vehicles and was assigned to the SUV category based on powertrain size. Light trucks represented 38% of the SUV category. Geometric constructions of the cabin models are shown in Figure 2. Once constructed, vehicle cabin model performance was validated with thermal soak and cooldown experimental data collected at NREL.

The cabin thermal models capture the solar intensity and angle relative to the vehicle throughout the day based on location and weather data. Model vehicle orientations are fixed relative to solar coordinates, and thermal loads are therefore dependent on vehicle orientation. To decrease the total computational time of the full analysis, a trial simulation was performed with the compact vehicle to determine which orientation should be used; thermal loads for the vehicle facing west most closely represented the four cardinal-direction average, agreeing to within 99.4%. The west orientation was therefore the selected orientation for the full national-level analysis. Additional vehicle cabin thermal model parameters are provided in Table 1.

The 20[degrees]C cabin cooling setpoint was set lower than a more common 22[degrees]C setpoint to account for some driver's lack of understanding of the HVAC control system and the tendency to reduce the setpoint to "Lo" or 15[degrees]C under the false assumption that this will accelerate the cooldown.

For temperatures less than 35[degrees]C, all outside air was assumed because that is how the authors thought many drivers run their manual systems to minimize odors and get fresh air into the cabin. For temperatures above 45[degrees]C, the authors assumed some drivers do not know how to properly control their HVAC efficiently and 50% of vehicles would not switch to recirc or have automatic recirc functionality.

Since the cabin thermal model and the A/C model were decoupled, a peak cooling capacity was defined for the cabin thermal model. The peak cooling capacity in Table 1 capped the evaporator capacity at the respective values. These values were determined through discussions with automotive HVAC industry experts. The SUV peak capacity assumes a dual-evaporator system; therefore, including pickup trucks with SUVs lightly overpredicts A/C fuel use for this platform.

Implementation of Solar/Thermal Technologies in Cabin Model

Thermal Control Glazing. Two glazing configurations were used in the analysis by selection of the direct solar transmittance (Tds) as defined in ISO 13837 [13] in the vehicle cabin models. For the baseline compact and mid-size configurations, the Tds for the windshield was 41.9% and the Tds for all other glazings was 45.1%. For the thermal control glazing configuration, the Tds for the windshield and all other glazings was 33.0%. For the SUV, a Tds of 16.6% was applied to the glazings behind the B pillar for both configurations, while the glazings in front of the B pillar matched the compact/mid-sized Tds values for both configurations. The thermal control glazing properties were based on a triple silver layer IR reflective configuration.

Solar Reflective Paint. Weighted average radiative properties were calculated for the baseline vehicle exterior paint by using PPG Industries color trend market analysis data [14] and the Lawrence Berkeley National Laboratory Pigment Database [15]. The baseline paint solar-weighted absorptance was 64.1% with 60.9% absorptance in the near-infrared spectrum. Solar reflective paint is defined in the 2017-2025 greenhouse gas (GHG) emissions regulation as having less than 35% absorptance in the near infrared. For this reason, the solar-weighted absorptance of the solar reflective paint configuration was calculated by reducing the baseline near-infrared absorptance to 35%, resulting in a solar-weighted absorptance of 48.3%.

Passive and Active Ventilation. The cabin air infiltration rates for the baseline vehicle configurations were 0.35, 0.45, and 0.7 air changes per hour for the compact, mid-size, and SUV, respectively, and were based on outdoor experimental data collected at NREL on multiple vehicles. A tracer gas decay method was implemented using two Bruel & Kjaer Multi-Gas Type 1302 Monitors. A small amount of SF6 was injected into two vehicles, and the concentrations were monitored for ~ 2 hours. Wind speeds were measured and data during periods of sustained winds ~ [greater than or equal to] 4 m/s were not used. Methods used for both passive and active ventilation are not explicitly defined in the 2017-2025 GHG emissions regulation, and technology implementation is therefore open for interpretation. In a previously conducted passive ventilation test strategy at NREL, an air temperature reduction of 3.5[degrees]C was measured due to lowering the door windows by 2 cm, increasing the air infiltration rate during a thermal soak experiment [16]. To determine the air infiltration rate for the passive ventilation configuration, the vehicle cabin thermal model infiltration rate was adjusted to match the cabin air temperature reduction obtained experimentally. An infiltration rate of 4.8 air change per hour during a thermal soak resulted in a 3.5[degrees]C temperature reduction and was used to represent passive ventilation.

In a previously conducted thermal load reduction test program at NREL, a just-in-time active ventilation strategy was found to be effective at reducing the soak temperature in an electric vehicle [17]. Using the HVAC blower to ventilate the cabin 15 minutes prior to vehicle operation, the average cabin air temperature was reduced by 7[degrees]C. To determine the air infiltration rate for the active ventilation configuration, the vehicle cabin model infiltration rate was adjusted to match the cabin air temperature reduction obtained experimentally. A 30 l/s (65-cubic feet per minute) per minute infiltration rate implemented 15 minutes prior to drive events was used to represent active ventilation in the three vehicle models. The energy to run the blower was not included in the calculation. Although the alternator would need to recharge the 12-volt battery upon engine start, the energy of the blower producing 30 l/s for 15 minutes was considered negligible compared to the A/C compressor energy for a drive.

Environmental Conditions and Vehicle Populations Light-duty vehicle registration data for each county within the United States was obtained through the Polk database [12]. In addition, Typical Meteorological Year weather data were obtained for over 900 weather stations located throughout the United States. A two-step process was used to associate vehicle populations with environmental conditions and reduce the number of simulated locations required in the analysis. First, county vehicle registrations were assigned to the geographically closest weather station based on the distance from the county center, omitting any weather station without assigned county registrations. Next, locations were sequentially eliminated based on the location with the smallest number of county registrations, with reassignment of the eliminated registrations to the nearest remaining location. In addition, at least one weather location per state was retained regardless of the number of registrations assigned, to maintain geographic coverage in low population density areas of the United States. The location elimination process was ceased when elimination of a location caused reassignment of 0.25% or more of the national vehicle registrations. Maps of the initial 839 U.S. weather locations and the remaining 206 locations after completion of the down-selection process are provided in Figure 3.

Vehicle Usage Patterns. Diurnal fluctuations in the environment strongly influence vehicle soak temperatures and thermal loads needed to cool the vehicle cabin. In addition, the duration of a drive event significantly impacts thermal loads with higher loads occurring early in the drive event. Finally, previous vehicle usage and associated short soak times can reduce the thermal loads necessary for subsequent drive events. For these reasons, information from the 2009 National Household Travel Survey (NHTS) [18] was extracted to determine light-duty vehicle travel behaviors. In the statistical analysis of the NHTS vehicle usage data, automobile, van, SUV, and pickup truck data were combined since the respective usage patterns were similar. The survey interface then applied weighting factors to calculate national vehicle usage data based on the survey locations. This usage was used for the three vehicle classes. Vehicle trip start time, soak durations, and trip duration were consolidated to provide representative behaviors and associated weighting factors. A distribution plot of light-duty vehicle trip start times extracted from the NHTS is provided in Figure 4. The survey data indicate significant driving concentrated throughout the daytime and into the evening. For the analysis, three representative vehicle trip start times of 7:06, 12:35, and 18:26 were used to represent the distribution, capturing morning, midday, and evening drive events and associated environmental conditions.

A distribution plot of vehicle trip durations extracted from the NHTS is provided in Figure 5. Survey data indicate a large percentage of light-duty trips are less than 15 minutes in duration and over 90% of trips are less than 45 minutes in duration. For the analysis, three representative trip durations of 7.2 minutes, 18.4 minutes, and 49.4 minutes were used to characterize the distribution, capturing short duration drives (less than 15 minutes), medium duration drives (from 15 to 30 minutes), and long drive events (30 minutes and greater).

Finally, the time duration that a vehicle is not operated between trips is vehicle dwell time or soak time. Data from the NHTS indicated the 50th percentile for dwell time was 50 minutes. For the analysis, two representative vehicle soak duration groups were selected to represent both short dwell time and long (full vehicle soak) dwell time scenarios. The first group represents short soak duration events and includes soak durations from 0 to 50 minutes with an average of 17 minutes. The second group represents long soak duration events and includes soak durations from 50 minutes to a full day with an average of 232 minutes. Combining the three start times, three drive durations, and two soak durations results in a total of 18 use simulation conditions. The usage conditions and corresponding weighting factors used in the analysis are summarized in Table 2.

Air Conditioning System Model

An A/C system model was configured to represent a 2007-2010 model year vehicle with R134a refrigerant, belt-driven 200-[cm.sup.3] fixed displacement compressor with a pulley ratio of 1.37. This type of system was selected to be similar to the data U.S. EPA and NHTSA used in the development of the regulation. In the model, refrigerant superheat was controlled by a thermal expansion valve. Evaporator condensate icing conditions were avoided by controlling the evaporator outlet air temperature to 3[degrees]C through clutch-cycling the compressor. The compressor was also cycled due to high- and low-pressure limits when necessary. A constant condenser fan power of 75 W, based on estimates in the Green-MAC-LCCP model [5], was included with the calculated compressor power. The HVAC blower was assumed to be operating regardless of A/C compressor operation, and a 150-W accessory load was added to the vehicle, representing a moderate operating condition. This load did not contribute to the A/C fuel use or off-cycle credit results. In addition, two control strategies were considered: ideal and automatic. In the ideal case, the A/C load perfectly matched the thermal load of the cabin. In the automatic case, A/C loads due to reheating HVAC operation for a given trip were estimated to be equal to the peak thermal load for the remaining duration of the trip after it occurred during the transient cooldown. The increased A/C loads due to reheating were estimated to occur 38% of the time the A/C system was operated based on assumptions from the EPA/NHTSA final rulemaking [2]. A/C operation for dehumidification was not included in the analysis.

National-level co-simulation using the detailed CoolSim A/C system model with the cabin model was prohibitive due to the simulation time necessary for the A/C system model, which operates on small time steps. Therefore, the two models were decoupled as previously described, and the A/C system performance was characterized for a range of input conditions, including condenser and evaporator inlet conditions, vehicle speed, engine speeds, and evaporator capacity (Table 3). This characterization was captured in an eight-dimensional lookup table that is available for future analyses. The A/C system performance for each cabin thermal load time step was then computed by interpolation of the characterized system. The decoupled approach introduced an approximation that instantaneous A/C system performance was equivalent to performance obtained using the precalculated data. To estimate the error due to this assumption, the dynamic A/C system performance of a select drive event was computed and compared to the precalculated performance method, with a difference of less than 10%. The error associated with the approximation was acceptable, considering prohibitive computational requirements for a co-simulated analysis and technology performance estimations determined through differential comparison. Since the compressor was characterized by volumetric, isentropic, and mechanical curves and the system evaporator capacity was capped in the cabin thermal model, the 200-[cm.sup.3] displacement of the compressor was not a key assumption. The system coefficient of performance (COP) was a mean value over periods of time that included inefficiencies introduced due to cycling the compressor.

The range of the parameters in Table 3 defines the edges of the eight-dimensional COP lookup table. If an input parameter exceeds the range, the nearest value is used, and extrapolation outside the limits is avoided. The evaporator air temperature upper limit is greater than the condenser to cover the case where hot cabin recirculation air is mixed with outside air.

Vehicle Fuel Use Model

Vehicle models were developed in FASTSim [9] to calculate the impact of the A/C system accessory load on vehicle fuel use and associated CO2 emissions for each of the three representative vehicle platforms. Data from NREL's Transportation Secure Data Center (TSDC) were used to evaluate and identify a representative drive cycle for each of the three representative drive durations [19]. All drive events with drive times similar to the target drive time were extracted from the TSDC, and aggregate statistics were computed for the three representative durations. Drive cycle statistics from NREL's TSDC are provided in Table 4. For each trip, deviation statistics were calculated, and a representative cycle was defined by selecting the cycle with the minimum composite deviation. Vehicle speed versus time plots for the selected vehicle drive cycles for the 7-, 18-, and 49-minute representative drives are provided in Figure 6. The representative drive cycles were used in the calculation of fuel use in the vehicle model and also applied to the cabin model wind speeds for the vehicle exterior.

Once representative drive cycles were selected, the vehicle models of the three vehicles were used to evaluate vehicle performance over a range of accessory loads for each of the cycles. A vehicle performance map was created with fuel use as a function of drive cycle, vehicle type, and accessory load.

Computational Process

For the national-level analysis process, a full factorial simulation was performed to incorporate representative national vehicle use cases. In addition to 206 weather data sets from the down-selected registration-weighted locations, the three vehicle platforms were simulated for three representative time-of-day of travel, three trip durations and associated representative drive cycles, and two representative vehicle soak durations. All simulations were completed for both the baseline vehicle configuration and the solar/thermal control technology vehicle configurations. Full factorial analysis resulted in 4,944 simulations for each of the three vehicle platforms, resulting in a total of 14,832 annual simulations with 1-minute time steps. Due to the large number of simulations necessary, a high-performance computing (HPC) system was used for parallel simulation [20]. The simulation results were aggregated using post-processing, and appropriate weighting factors were applied to incorporate the relevance of each use case.

Results

Impact of A/C Control, Vehicle Usage Patterns, Vehicle Platform, and Environment

The impact of the A/C control strategy on baseline A/C fuel use is shown in Figure 7. Assuming annual vehicle miles traveled of 11,346 miles/year [21], the baseline A/C fuel use with ideal control was 27.5 gal/yr/vehicle for the three-vehicle platform average. This assumes all the miles traveled with A/C on used this control strategy. The automatic control strategy resulted in a higher A/C fuel use of 34.2 gal/yr/vehicle.

Figure 8 shows the impact of the A/C control strategy on the reduction in A/C fuel use due to the four solar/thermal technologies. The percent savings were calculated using the baseline A/C fuel use in Figure 7 for the respective control strategies. Since the solar control glass and paint reduced the thermal load throughout the entire drive, the ideal control strategy took advantage of the lower thermal loads and realized larger savings. While the solar control glass savings for the automatic temperature control strategy were still significant, some of the benefit was not captured due to the overcooling-reheat assumption. Since each automotive manufacturer has a unique approach to HVAC automatic control, our analysis process attempted to assess the upper and lower bounds. Since the extreme low and high energy use cases simulated here did not significantly change the overall baseline A/C fuel use and percent fuel savings for each technology, the details on how automatic temperature control is implemented are less critical.
FIGURE 7 Impact of A/C control strategy on baseline A/C fuel use.

Ideal Control                  27.5
Automatic Temperature Control  34.2

Note: Table made from bar graph.


The impact of trip duration is shown in Figure 9. As in the previous example, the vehicle miles traveled were assumed to be 11,346 miles/year, and all miles were replications of the respective trip durations. Short trips had a significantly larger baseline A/C fuel use because more time was spent in the transient cooldown mode where the thermal loads are higher. Longer trips had more steady-state A/C operation and lower A/C fuel use when the thermal loads were lower. Figure 10 shows the impact of trip duration on the percent A/C fuel use savings for the four solar thermal technologies. The active and passive ventilation were relatively insensitive to trip duration. Solar control glass and paint had larger percent reductions in A/C fuel use as the trip duration increased because these technologies reduced the thermal loads in the steady-state as well as transient portions of the drive.
FIGURE 9 Impact of drive duration on baseline A/C fuel use.


Short Trip   33.4
Medium Trip  28.9
Long Trip    22.7

Note: Table made from bar graph.


Figure 11 shows the impact of the soak duration on baseline A/C fuel use. On days when there was solar radiation present, the vehicle interior was initially warmer for the full soak case than the partial soak case, resulting in a baseline A/C fuel use of 32.0 gal/yr for the full soak case. While the interior temperature rises rapidly when a solar radiant load is present, the interior temperatures at the start of the drive after the 17-minute partial soak are lower than the full soak and result in a lower baseline A/C fuel use of 28.1 gal/yr.
FIGURE 11 Impact of soak condition on baseline A/C fuel use.

Full Soak     32.0
Partial Soak  28.1

Note: Table made from bar graph.


The percent A/C fuel use savings for the four solar/thermal technologies are shown in Figure 12. There was no parked car ventilation for the partial soak because the vehicle interior started cooler than ambient and this strategy avoids pulling hot outside air into the cool interior. For the solar control glass and paint, the larger savings for the partial soak may be due to these technologies effectively reducing the rapid rise in interior air temperature from the cold initial condition.

Figure 13 shows the baseline A/C fuel use as a function of start time. The midday start time had the highest A/C fuel use because this is when the interior temperatures and solar load were the highest. In contrast, the morning drive had significantly lower baseline A/C fuel use because the thermal loads were significantly lower due to the negligible solar soak and cool morning temperatures. Figure 14 shows the interior air temperatures for a mid-sized vehicle with a long drive and full soak on September 2 in Miami, Florida. The simulation included a morning, midday, and evening drive. Note the high soak temperature just prior to the drive at noon and the significant reduction due to the solar control glass. During the noon drive, the solar control glass resulted in lower cabin air temperatures, although neither baseline or solar control glass configurations attained the 20[degrees]C setpoint temperature. In the morning drive, solar control glass had no impact because vehicle operation occurred prior to a significant solar load. In this low thermal load situation, the solar control glass and baseline configurations had the same cabin soak temperature and attained the 20[degrees]C setpoint temperature quickly for the duration of the drive.
FIGURE 13 Impact of drive start time on baseline A/C fuel use.

Morning Drive   6.8
Midday Drive   42.0
Evening Drive  25.9

Note: Table made from bar graph.


Figure 15 shows the baseline A/C fuel use by month for the three start times for the mid-sized vehicle. The midday drive had the highest A/C fuel use in all months and peaked in July. The morning drive had the lowest A/C fuel use in all months and was zero in the winter months.
FIGURE 16 Impact of vehicle platforms on baseline A/C fuel use.

Compact  27.9
Midsize  29.0
SUV      31.4

Note: Table made from bar graph.


Figure 16 shows the impact of vehicle platform on baseline A/C fuel use. The compact vehicle had the lowest A/C fuel use of 27.8 gal/yr, while the SUV had the largest at 31.4 gal/yr. This difference is primarily a function of cabin geometry, vehicle properties, and the maximum capacity of the A/C system.

As in the previous driver-use cases and HVAC-control cases, the percent savings in A/C fuel use for active and passive ventilation are low for all three vehicle platforms in Figure 17. For solar control glass, the mid-sized vehicle had the largest percent reduction in A/C fuel use because the sloped windshield and backlite caused this configuration to be the best solar collector (higher ratio of glass surface area to total surface area). The SUV had the lowest percent savings because solar control glass was not applied to the darkly tinted glass behind the B pillar. For the solar reflective paint, the mid-sized vehicle had the lowest percent reduction in A/C fuel use because this platform had the lowest painted surface area to total surface area ratio. This is consistent with the solar glass result and the higher ratio of glass surface area to total surface area.

Figure 18 shows the influence of environmental conditions on A/C fuel use. As expected, states with a hot and humid environment had higher A/C fuel use. Hawaii had the largest A/C fuel use at 70.8 gal/yr, and Alaska had the lowest at 4.5 gal/yr. For states with multiple weather locations, the calculated average value had equal weighting.

Reviewing the sensitivities of the simulated parameters, the environment, time of drive, and trip duration had the largest impact on A/C fuel use. This makes the associated weighting factors and assumptions more important for these parameters. HVAC control, soak duration, and vehicle model had a lesser impact, and therefore the associated assumptions were less critical.

National-Level Results

A contour plot of the baseline annual U.S. light-duty A/C fuel use for the weighted average of the three vehicle platforms is provided in Figure 19. As expected, A/C fuel use shows a strong dependence on locations throughout the United States with larger fuel consumption in higher populated areas with warmer/higher humidity climates. Increased A/C fuel use in the warmer locations was due to both higher peak loads and increased use throughout the year. The U.S. light-duty fleet A/C fuel use was calculated to be 7.59 billion gallons of gasoline per year based on a U.S. fleet size of 252,714,871 vehicles [12] and each traveling on average 11,346 miles/year [21]. This is equivalent to 6.1% of the total U.S. light-duty vehicle fuel use based on annual U.S. light-duty vehicle fuel use of 123.6 billion gallons [22] and is similar to previous NREL estimate of 5.5% [3].

In addition, the A/C fuel use per vehicle is shown in Figure 20. This plot eliminates the influence of vehicle population and emphasizes the impact of environmental conditions showing that warm and humid locations have the largest A/C fuel use. The average vehicle A/C fuel use was calculated to be 30.0 gallons of gasoline per year. These results compare closely to the results of the previous NREL analysis of 7.0 billion gallons per year and 30.8 gallons/yr per vehicle [3]. Implementing advanced technologies to reduce vehicle thermal loads at the national level can potentially save millions of gallons of fuel, reducing oil consumption for light-duty vehicle transportation.

In addition to baseline A/C fuel use, annual fuel use was calculated for the four solar/thermal control technologies previously described. Table 5 shows the national and per-vehicle reductions in A/C fuel use for the solar/thermal control technologies compared to the baseline configuration assuming 100% adoption for the technologies. Solar control glass had the largest reduction in A/C fuel use, saving 661 million gallons per year, or an average vehicle savings of 2.62 gallons of gasoline per year. Solar reflective paint had the second largest reduction in A/C fuel use (180 million gallons), followed by passive ventilation (71.3 million gallons), and finally active ventilation (42.2 million gallons).

The impact of both solar control glass and solar reflective paint was significant and expected due to the ability of the technologies to reduce transient and steady-state thermal loads during both soak and drive events. In contrast, both active and passive ventilation had a smaller impact on A/C fuel use, with total fuel savings of 42,200,000 and 71,300,000 gallons, respectively (average vehicle savings of 0.17 and 0.28 gallons of gasoline per year). While the solar control glass and solar reflective paint reduced thermal loads for the entire drive, both active and passive ventilation did not contribute to the reduction in vehicle thermal loads during the steady-state portion of the drive event, but instead reduced the peak soak temperatures in the vehicle prior to the drive. Another contributing factor to the reduced impact of ventilation was that 50% of the drives were considered to be partially soaked (a predrive followed by a 17-minute soak). Since the cabin interior started the short soak period from a cooled condition in this case, pre-ventilation was not used because the ambient air would be warmer than the interior. In addition, implementation details for the two ventilation strategies are not specified in the regulation; therefore, their performance will depend on the individual manufacturer's design, which will be different than what was assumed in this analysis. Ventilation can be effective at reducing the cabin air temperature immediately prior to the drive event; however, stored energy in the interior components and walls may still remain, limiting the effectiveness of the technology. Therefore, while ventilation can be effective at relieving thermal discomfort of the occupants early in the drive, its impact on reducing the transient thermal loads necessary to attain a target vehicle cabin temperature is attenuated. Finally, in this study the reduction in fuel use due to passive ventilation was higher than for active ventilation. While the exact cause has not been determined, it is likely due to implementation strategy differences. In the active ventilation strategy, a just-in-time control strategy was implemented, providing ventilation 15 minutes prior to the drive event, in contrast to passive ventilation that was present throughout the entire day when the vehicle was not in operation. Varying ventilation strategies and flow rates are expected to impact the effectiveness of extracting stored energy from the vehicle interior. This result highlights the need to provide well-defined ventilation strategies to maximize their effectiveness.

After A/C fuel use estimations were calculated for both the baseline vehicle and solar/thermal control technologies, the associated C[O.sub.2] emissions were calculated based on an estimated 8,887 grams of C[O.sub.2] per gallon of gasoline [23]. The results of the calculated C[O.sub.2] emissions for the vehicle configurations are provided in Table 6. The national average annual light-duty A/C C[O.sub.2] emissions are 23.5 grams per mile. This is higher than 11.9 and 17.2 grams per mile for cars and trucks, respectively, that the EPA calculated in support of the U.S. Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards. Active and passive ventilation were determined to have a relatively small impact with 0.1 and 0.2 gram per mile savings, respectively, for reasons discussed in the previous paragraph. However, solar control glass and solar reflective paint had a significant reduction in C[O.sub.2] emissions, at 2.0 and 0.8 grams per mile, respectively. Emissions reductions for passive and active ventilation were significantly lower than the EPA values. The emissions reductions for solar control glass and solar reflective paint showed reasonable agreement with the EPA values.

Conclusions

NREL developed a rigorous analysis process to calculate vehicle A/C fuel use and used it to simulate the impact of solar/thermal control technologies. The process used the software developed under previous U.S. Department of Energy Vehicle Systems projects: CoolSim, CoolCalc, and FASTSim. The process accounts for vehicle use patterns, vehicle type, environment conditions, vehicle population, and A/C system efficiency. The analysis assumptions were selected to be consistent with vehicle model year and technology when the GHG regulation was developed to enable comparison to the off-cycle GHG solar/thermal credits. Consequently, the A/C fuel use results are applicable from ~ 2007 to 2010 timeframe. The results show that the United States uses 7.6 billion gallons of fuel annually for vehicle A/C, which is equivalent to 6.1 % of the national light-duty vehicle fuel use. This significant contribution of vehicle A/C to national fuel use presents an opportunity for advanced thermal control technologies to reduce U.S. fuel consumption, reduce imported oil, and improve energy security.

On a per-vehicle level, the analysis showed that the A/C system operation increases A/C fuel use by 30 gal/yr/vehicle, or 23.5 g of CO2 per mile. This compares to the 11.9 and 17.2 g/mi for cars and trucks, respectively, that the EPA calculated in support of the U.S. Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards. Since the EPA used these values to calculate off-cycle credits and caps, there is an opportunity for increased off-cycle credits and caps if updated, larger baseline C[O.sub.2] emissions per mile were used.

Four thermal control technologies (solar reflective glass, active ventilation, passive ventilation, and solar reflective paint) identified in the off-cycle solar/thermal credit menu in the Final Rule for Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards were simulated. On a per-vehicle basis, solar reflective glass reduced fuel use by 2.62 gal/yr and C[O.sub.2] emissions by 2.0 g/mi. This compares to the EPA off-cycle credit of up to 2.9-3.9 g/mi. Solar reflective paint is predicted to reduce CO2 emissions by 0.8 g/mi (1.0 gal/yr), while the EPA credit is 0.4-0.5 g/mi. These emissions reductions showed reasonable agreement with EPA values. Active and passive ventilation had a lower savings of 0.1 and 0.2 g/mi, respectively, compared to the off-cycle credit values of 2.1-2.8 g/mi for active ventilation and 1.7-2.3 g/mi for passive ventilation. NREL's ventilation results were lower compared to the EPA credits due to NREL's definition of ventilation strategies, the use of drive cycles that include a significant steady-state component, thermal energy of the vehicle interior mass, and 50% of the drives not using ventilation because the interior temperature is less than the ambient temperature. These technologies and other advanced thermal concepts can reduce A/C fuel use and C[O.sub.2] emissions while reducing the amount of oil imported into the United States.

Contact Information

John P. Rugh

National Renewable Energy Laboratory

15013 Denver West Parkway Golden, CO 80401

john.rugh@nrel.gov

Jason Lustbader

jason.lustbader@nrel.gov

Acknowledgments

This work was supported by the U.S. Department of Energy under Contract No. DE-AC36-08GO28308 with Alliance for Sustainable Energy, LLC, the manager and operator of the National Renewable Energy Laboratory. Funding was provided by the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Vehicle Technologies Office. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.

Definitions/Abbreviations

A/C - air conditioning

C[O.sub.2] - carbon dioxide

EPA - Environmental Protection Agency

FASTSim - Future Automotive Systems Technology Simulator

HPC - high-performance computing

HVAC - heating, ventilating, and air conditioning

NHTS - National Household Travel Survey

NHTSA - National Highway Traffic Safety Administration

NREL - National Renewable Energy Laboratory

Tds - direct solar transmittance

TSDC - Transportation Secure Data Center

References

(1.) "2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards," 77 Fed. Reg. 62624 (Oct. 15, 2012).

(2.) "Joint Technical Support Document: Final Rulemaking for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards," United States Environmental Protection Agency and National Highway Traffic Safety Administration, Aug. 2012.

(3.) Rugh, J., Hovland, V., and Andersen, S., "Significant Fuel Savings and Emission Reductions by Improving Vehicle Air Conditioning," Mobile Air Conditioning Summit, Washington, DC, Apr. 14-15, 2004.

(4.) Kambly, K. and Bradley, T., "Estimating the HVAC Energy Consumption of Plug-In Electric Vehicles," Journal of Power Sources 259(2014):117-124, 2014.

(5.) Papasavva, S., Hill, W., and Brown, R., "GREEN-MACLCCP [R]: A Tool for Assessing Life Cycle Greenhouse Emissions of Alternative Refrigerants," SAE Int. J. Passeng. Cars-Mech. Syst. 1(1):746-756, 2009, doi:10.4271/2008-01-0828.

(6.) Lustbader, J., Kreutzer, C., Kekelia, B., Jeffers, M. et al., "VTCab, Rapid Vehicle HVAC Load Estimation Tool," DOE Vehicle Technology Office, Vehicle Systems Annual Report, 2015, 417.

(7.) Lustbader, J., Kreutzer, C., Jeffers, M., Adelman, S. et al., "Impact of Paint Color on Rest Period Climate Control Loads in Long-Haul Trucks," SAE Technical Paper 2014-01-0680, 2014, doi:10.4271/2014-01-0680.

(8.) Kiss, T. and Lustbader, J., "Comparison of the Accuracy and Speed of Transient Mobile A/C System Simulation Models," SAE Int. J. Passeng. Cars - Mech. Syst. 7(2):739-754, 2014, doi:10.4271/2014-01-0669.

(9.) Brooker, A., Gonder, J., Wang, L., Wood, E. et al., "FASTSim: A Model to Estimate Vehicle Efficiency, Cost and Performance," SAE Technical Paper 2015-01-0973, 2015, doi:10.4271/2015-01-0973.

(10.) Lustbader, J.A., Rugh, J.P., Rister, B.R., and Venson, T.S., "CoolCalc: A Long-Haul Truck Thermal Load Estimation Tool," SAE Technical Paper 2011-01-0656, 2011, doi:10.4271/2011-01-0656.

(11.) "National Solar Radiation Data Base: Typical Meteorological Year 3," http://rredc.nrel.gov/solar/old_data/nsrdb/1991-2005/tmy3/.

(12.) "2014 Polk Vehicle Registration Database," IHS Automotive, driven by Polk, https://www.ihs.com/btp/polk.html.

(13.) "Road Vehicles-Safety Glazing Materials-Method for the Determination of Solar Transmittance," ISO 13837:2008(E), First Edition, Apr. 14, 2008.

(14.) "Automotive Color Trends," PPG Industries, http://corporate.ppg.com/Color/Color-Trends/Automotive-Color-Trends.aspx, accessed 7/2016.

(15.) "Lawrence Berkeley National Laboratory Pigment Database," Lawrence Berkeley National Laboratory, http://coolcolors.lbl.gov/LBNL-Pigment-Database/database.html, accessed 7/2016.

(16.) Rugh, J. and Farrington, R., "Vehicle Ancillary Load Reduction Project Close-Out Report: An Overview of the Task and a Compilation of the Research Results," Golden: National Renewable Energy Laboratory, NREL/TP-540-42454, 2008.

(17.) Jeffers, M., Chaney, L., and Rugh, J., "Climate Control Load Reduction Strategies for Electric Drive Vehicles in Warm Weather," SAE Technical Paper 2015-01-0355, 2015, 2015, doi:10.4271/2015-01-0355.

(18.) "National Household Travel Survey, 2009," Oak Ridge National Laboratory, http://nhts.ornl.gov, accessed 5/2016.

(19.) "Transportation Secure Data Center," National Renewable Energy Laboratory, http://www.nrel.gov/transportation/secure_transportation_data.html, accessed 7/2016.

(20.) "Information for Users of the WINHPC System," National Renewable Energy Laboratory, https://hpc.nrel.gov/users/systems/winhpc, accessed 5/2016.

(21.) "Annual Vehicle Distance Traveled in Miles and Related Data," Highway Statistics 2013, U.S. Department of Transportation Federal Highway Administration.

(22.) Ward's Automotive Yearbook, 2014, http://wardsauto.com/wards-automotive-yearbook.

(23.) United States Environmental Protection Agency, "Greenhouse Gas Emissions from a Typical Passenger Vehicle," EPA-420-F-14-040a, May 2014.

John Palmer Rugh, Cory Kreutzer, Bidzina Kekelia, Gene Titov, and Jason Lustbader, National Renewable Energy Laboratory, USA

(*) 2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards," 77 Fed. Reg. 62624 (Oct. 15, 2012).

History

Received: 06 Mar 2018

Revised: 31 Jul 2018

Accepted: 11 Sep 2018

e-Available: 11 Dec 2018

doi:10.4271/06-12-01-0002
TABLE 1 Select vehicle cabin thermal model parameters used in analysis.

Cabin cooling       20[degrees]C
setpoint
Air recirculation   [T.sub.amb] < 35[degrees]C            0
fraction            35[degrees]C [less than or equal to]  Linear: 0-0.5
                    [T.sub.amb] [less than or equal to]
                    45[degrees]C
                    [T.sub.amb] > 45[degrees]C            0.5
Cabin volume [m3]   Compact                               2.4
                    Mid-size                              3
                    SUV                                   4.3
Peak cooling        Compact                               7.0 kW
capacity [kW]       Mid-size                              8.0 kW
                    SUV                                   9.0 kW

TABLE 2 Vehicle usage pattern group representations used in the
analysis and associated weighting factors.

Trip start time       Group 1     Group 2      Group 3

Time range             0:01-9:00   9:01-16:00  16:01-24:00
Representative trip    7:06       12:35        18:26
start time
Weighting factor       0.183       0.476        0.341

Trip duration         Group 1     Group 2      Group 3

Duration range [min]   0-15       15-30        > 30
Representative trip    7.2        18.4         49.4
duration [min]
Weighting factor      0.508        0.310        0.182

Soak duration         Group 1             Group 2

Duration range [min]   0-50             50-1,440 (full day)
Representative soak   17               232
duration [min]
Weighting factor       0.5               0.5

TABLE 3 Input parameters for the A/C system lookup table.

Parameter                                  Range of values

Engine speed [rpm]                           800, 1,400, 2,000,
                                           2,500, 3,000
Evaporator inlet temperature [[degrees]C]     15, 32.5, 50
Evaporator inlet relative humidity [%]        20, 50, 80
Condenser inlet temperature [[degrees]C]      15, 30, 45
Condenser inlet relative humidity [%]         20, 50, 80
Vehicle velocity [m/s]                         0, 12, 26
Evaporator capacity [W]                    1,000; 3,000; 5,000;
                                           8,000

TABLE 4 Calculated drive cycle statistics extracted from NREL's TSDC
for each of the three representative drive times.

Vehicle trip               Short        Medium      Long

Trip duration [min]            6.9-7.2   18.1-18.6   49.1-49.6
Cycles evaluated           2,243        860         101
Avg. distance [miles]          2.6        8.5        28.4
Avg. idle [%]                 17.0       18.9        14.5
Avg. driving speed [mph]      25.3       33.2        39.6
Avg. acceleration [mph/s]      1.1        0.9         0.8

TABLE 5 National per vehicle A/C fuel use and reductions in A/C fuel
use for solar/thermal control technologies.

                                            Reductions due to solar
                                               /thermal control
                                                 technologies
Solar/thermal control   Average vehicle     Average     U.S. light-duty
technology              national A/C        vehicle     fleet savings
                        fuel use            savings     [gal/year] (*)
                        [gal/year/vehicle]  [gal/year]

Baseline                30.0                N/A         N/A
Active ventilation      29.9                0.17         42.2 million
Passive ventilation     29.7                0.28         71.3 million
Solar control glass     27.4                2.62        661 million
Solar reflective paint  29.0                1.00        180 million

(*) Based on U.S. light-duty vehicle fleet size of 252,714,871 vehicles
[12], individual vehicles traveling 11346 miles/year [21].

TABLE 6 Direct C[O.sub.2] emissions for vehicle configurations
evaluated, including expected emissions savings for the solar/thermal
control technologies and the EPA credit values.

                        Individual   Individual  EPA car:    EPA truck:
Vehicle configuration   vehicle A/C  vehicle     baseline    baseline
                        C[O.sub.2]   savings     emissions   emissions
                        emissions    [g/mi]      due to A/C  due to A/C
                        (**) [g/mi]              and credit  and credit
                                                 [g/mi]      [g/mi]

National Baseline       23.5                          11.9        17.2
Vehicle
Active ventilation      23.4         0.1               2.1         2.8
Passive ventilation     23.3         0.2               1.7         2.3
Solar control glass     21.5         2.0         Up to 2.9   Up to 3.9
Solar reflective paint  22.7         0.8               0.4         0.5

(**) Based on 8,887 grams of C[O.sub.2] per gallon of gasoline [23].
COPYRIGHT 2019 SAE International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rugh, John Palmer; Kreutzer, Cory; Kekelia, Bidzina; Titov, Gene; Lustbader, Jason
Publication:SAE International Journal of Passenger Cars - Mechanical Systems
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2019
Words:7843
Previous Article:PSO-Fuzzy Gain Scheduling of PID Controllers for a Nonlinear Half-Vehicle Suspension System.
Next Article:Enhanced Coil Spring Modeling in Passenger Car Suspension for Improved Target Setting Process.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters