Presentation on theme: "Drive Cycle Development and Real-world data in the United States"— Presentation transcript:
1Drive Cycle Development and Real-world data in the United States Working paper No. WLTP-02-17GRPE Informal Group on WLTP,2nd Meeting January 2009,Agenda Item 6.Drive Cycle Development and Real-world data in the United StatesEdward NamUS Environmental Protection AgencyWLTP meeting, Geneva, SwitzerlandJanuary 15, 2009
2Outline Drive cycle development history in US FTP & off-cycleInventory cyclesOther cycle developmentSAFD vs Vehicle Specific PowerNew driving activity dataUS EPA 5-cycle fuel economy labeling (drive cycle weighting)Next stepsApologies for the American units!
3Why drive cycles are important Serve as a standardized measurement stick for emissions and fuel economyCan compare across vehicles (benchmark)Manufacturers design vehicles to meet standards set by cycles and test proceduresServes as proxy for “average” or typical drivingEmissions standards are strongly dependent on the cycle and test procedureDrive cycles also change over time, with infrastructure, policy (speed limits), and technology (power:weight)This presentation focuses mainly on the drive cycle aspect of test procedure harmonization.
4FTP cycle development US Federal Test Procedure LA4 (City, UDDS) Developed in the late 1960’s to describe typical driving (acquired in Los Angeles)The highway HFET cycle was developed to describe a typical rural route (acquired outside Ann Arbor, MI in the 1970s)These 2 cycles used to describe fuel economy in the USUS06 (aggressive) and SC03 (A/C) cycles developed based on “3 cities data” of instrumented vehicles from Baltimore, Spokane, Atlanta.REP05, REM01, SC03 were developed to cover the full range of driving, but were simplified to US06 and SC03, while FTP remained in placeThese cycles are extreme to prevent “cycle beaters”
5Inventory Cycles - California Implemented new base cycleLA92 – determined from driving in LA in early 1990’s (40kph)More representative of 1990’s drivingFacility Cycles (speed correction factors)12 unified correction cycles (UCC) mainly from chase car data in LAFrom 4 to 95kphChosen by mean speed, speed-acceleration frequency distribution, positive kinetic energy (PKE), load, maximum acceleration, maximum deceleration, percent idle, percent acceleration, distance, etcThese cycles are used to correct the base emission factor from LA92 to other speeds
7US EPA Inventory Cycles USEPA - Facility cycles (1997)Developed by Sierra Research from 3 cities chase car data (Baltimore, Spokane, LA)11 cycles based on roadway type and congestion level (+ramp)Each cycle lasts ~10 minutesMatched second-by-second segments of chase car data by comparing SAFDs (speed acceleration frequency distribution)Can find on EPA website under MOBILE6 technical support documentation
8Speed/Acceleration Frequency Distribution (SAFD) A typical way to represent the overall driving behavior is a speed/acceleration frequency distribution. However, comparing SAFDs is difficult, since the relative impact of each bin on emissions can vary.Normally, the effect of these differences could only be properly quantified by expensive laboratory testing of vehicles. This is what was done to address the vehicle activity observed in the 1992 studies.
9EPA & Sierra cycle specifications time in acceleration/decelerationtime at cruisetime at idlemaximum speedaverage speedaverage or predominant speed during cruisemaximum acceleration/deceleration ratemaximum powerlength (time and miles)stops per mileaverage positive kinetic energy (PKE) change per mile and specific powerdistributions of speed and acceleration
10USEPA Cycle development (cont’d) Microtrips were chosen based on how well they matched the specificationsCycle choice criterialowest sums of differences on SAFDmatching real world power (2va)Segments shortened or lengthenedCycles used for (emissions) “speed correction factors” and for speed dependent transportation planning in MOBILE6 (emissions inventory model)
11UC-Riverside CE-CERT International Vehicle Emission Model data collection and cycle development for rapid emissions inventory estimationAlmaty, KazakhstanBeijing, ChinaLima, PeruMexico City, MexicoNairobi, KenyaPune, IndiaSantiago, ChileSao Paulo, BrazilShanghai, China
12Vehicle Specific Power: an alternate metric for cycle manipulation MOVES is EPA inventory model replacing MOBILEMOVES is a modal model based on VSP activityCycle metric should be based on a more physically causal variable for emissions formation: e.g.Road load (tractive) PowerP ~ Av + Bv2 + Cv3 + Mva*A,B,C are vehicle target coastdown coefficients* including road gradeVSP = P/M (M is mass of vehicle)Divided by mass since emissions (measured in g/km) is largely independent of vehicle massWith VSP distributions and proper modal data, emissions can be converted from one drive cycle to anotherThis approach has been validated by a number of studies
13VSP Distribution in MOVES 0-25 mph0-40 kph25-50 mph40-81 kph>50 mph, >80kphOne could make finer scale bins if necessarykW/tonne
14Cycle Development requires data Much second by second data has already been collectedBut these were not collected with harmonization in mind, so data is scattered and inconsistentRequire new and more rigorous analysis methods
15Real world driving data since 3 cities & LA92 Chase car:Los Angeles 2000Instrumented Vehicle:US EPA Ann Arbor shootout data 2001Kansas City dataAtlanta (Georgia Tech)1600+ vehicles, 800+ households, GPS, accelerometer, OBDVehicles instrumented for 2-3 yearsMost comprehensive activity data in existenceData excellent source for start activity as wellDue to privacy concerns, data has a finite lifetime
16Los Angeles Comparison Speed Los Angeles is a good place to start in comparing changes in vehicle activity. LA was both a site observed in 1992 and it is also a site in a more recent chase car study done in 2000.As expected, new (higher) speed limits have affected the average speed driven by vehicles.But is this really the full picture of the changes in vehicle activity?More time is spent at high speeds in 2000 than in 1992.Speed limits were increased during this period
17Los Angeles Comparison Acceleration More time is spent at high acceleration in 2000 than in 1992.Another way to compare vehicle activity is in terms of accelerations.There are more high accelerations in the new data than observed in 1992.But is this change significant, in terms of the effect on overall emissions from vehicles?
18Operating Modes for MOVES VSP bins For coast and cruise,13 modes retainedfrom MOVES2004, plus8 modes added forMOVES2006 formerly bins26 and 36PLUSOne mode each foridle, anddecel/brakingGives a total of23 opModes
19Los Angeles in 1992 compared to 2000 So, using the operating mode bins defined by MOVES2004, we can again compare the 1992 and 2000 calendar year chase car measurements in Los Angeles.Because VSP has a more direct and quantifiable relationship to vehicle emissions, these comparisons can more easily determine significant differences that will affect overall vehicle emissions.In this chart, note that the activity (time in mode) for calendar year 2000 is higher in the most significant bins for emissions (high speeds and high accelerations, bins 26 and 36).This suggests that the vehicle activity used in current models from 1992 is still not aggressive enough (bins 16, 26 and 36) to account for current (2000) vehicle behavior.
20Other Factors: California (2000) Urban vs Rural Another interesting factor in the new data is the California chase car study done outside of urban areas. Nearly all driving studies have been done in urban areas, since that is where the air quality problems usually occur and the majority of emissions are produced. This allows for the first time some comparison of vehicle driving behavior outside of urban areas.This new data will allow us to better account for driving activity outside of urban areas.Rural driving is faster than urban driving
21SouthEast Michigan Shootout data 15 PEMS instrumented vehicles.Driven by US EPA or SENSORS employees in calendar year 2001.Note: Drivers were not randomly chosen.Avg. distance driven per vehicle : 52 milesAverage speed : 31.2 mphSimilarly, 15 vehicles were instrumented in Michigan as part of the development of the MOVES2004 methodology.
22Kansas City (Round 1.5 Only) US EPA studyInstrumented vehicles (not chase car) from random population of newer vehicle ownersPEMS (Portable Emissions Measurement System) equipped to measure emissions and activity.Drivers measured for 15,000-30,000 seconds (battery lifetime)Measured conventional as well as hybrid vehiclesAvg. distance driven per vehicle: 41 milesAverage speed: 30.1 mphThere are other studies available other than the chase car study in Los Angeles.In Kansas City, vehicles were equipped with instrumentation to measure both vehicle activity and emissions.Vehicles were not driven during the entire measurement time (4 to 8 hours), which resulted in about 1.4 hours of driving per vehicle.This data is from round 1.5 only of the KC study, which are SFTP certified Tier 1 vehicles ONLY. The vehicles were selected randomly from that pool with the exception that an effort was made to specifically recruit a certain number of hybrids. The study includes 61vehicles, 21 of which were hybrids. This study was an EPA study only, designed primarily to quantify real-world fuel economy. The KC study is STILL ongoing and will be complete by the end of April, so the activity results for the city may change. The results should be considered preliminary and not a measure of total KC activity as recorded by rounds 1, 1.5 and 2. Those results are pending.
23Speed acceleration frequency distributions Based on a 2-dimensional matrix of speed and accelerationsFTP HFET is insufficient to describe real-world driving.Even though KC seems to be less aggressive than the US06 overall, it seems (as the next slide will hint) that LA driving may be as aggressive as the US06. This means that a cycle that was once designed to be extreme may now be more representative of real-world driving!
24Recent Activity Surveys When we compare all of these sources, all of the recent studies, when compared to MOVES2004, are more aggressive.MOVES2004 is based on the older 1992 data.Idle have been removed from the comparison, since instrumented vehicles record much more idle time than chase car studies (such as done in LA). This idle time must be accounted for properly, but is not as important as a vehicle activity issue.LA seems to have the most aggressive driving of the regions studiedRecent surveys have the more high speed, high power driving.LA driving is the most aggressive, & may be comparable to US06
25Conventional vs Hybrid Activity Hybrid drivers may drive less aggressively as a group, or their vehicles may be weaker powered in general – it’s probably a combination of these factors that gives these results25–50 mph>50 mph<25 mphHybrid driving is slightly less aggressive than conventional vehicles
265-Cycle fuel economy labeling Previously, label fuel economy was based on a 2 cycle number: city and highwayReal-world fuel economy effects were captured with a 20% correction factorThesis: Real-world driving can be described by a linear combination of existing drive cyclesIf weighted properly the activity and fuel economy can be captured better than a fixed correction factorThe key is to use cycles with a broad enough range to capture the VSP profileUS driving activity is mainly represented by a linear weighting of 3 cycles: FTP(city), HFET, and US06
28Other relevant 5 cycle issues Through a combination of FTP (city), highway, US06, cold FTP, and hot SC03 (air conditioning) real-world driving was bracketedResult: fuel economy label became more representative of what vehicle owners would truly get throughout a year of drivingPotential lessons for harmonization:VSP weighting methodology can be a powerful toolMay be able to represent different regions through a combination (and proper weighting) of drive schedules
29Comparison of FTP, NEDC, JC08 The cycle as it looks immediately after start will define when the vehicle goes closed loop, so how the modules are ordered could have a significant impact on fuel economy.The NEDC and JC08 have similar VSP profiles – in hot start, should have similar emissionsMain difference likely due to sequence following starts
30Conclusions US has a long history of drive cycle development Much new data exists in the US for a harmonization project (with more coming)Driving has been getting faster and more aggressive with each passing decadeVSP activity is a powerful (yet simple) tool to characterize and combine drive cycles and to compare driving from different regions
31Issues for future consideration Definition of “city” vs “highway” (urban vs extra-urban)City can include high speed driving and highways can be congested stop and goStarts/km variabilityShift schedulesScaling drive cycle for small engine “cars”
32Next stepsWe should analyze the existing data especially the Atlanta data before it is destroyedShould agree on QA/QC procedures for dataAs well as a methodology for cycle generation and comparing “representativeness”The US EPA is going to start collecting second by second activity and emissions data from real-world operation from a variety of cities
34Data Limitations OBD data is noisy when used for accelerations GPS is noisier than OBDShould make an effort to determine best filtering mechanism compared to directly measured accelerationsAccelerometer, 5th wheel, etc.Do accelerations matter?Certification tests have x% eror band for drivers to follow cyclesBut Robot drivers and computer simulations can follow cycles exactly, so the exact cycle trace can matter