Project at a Glance
Title: Characterization of driving patterns and emissions from light-duty vehicles in California.
Principal Investigator / Author(s): Austin, Thomas A
Contractor: Sierra Research, Inc.
Contract Number: A932-185
Research Program Area: Emissions Monitoring & Control
Topic Areas: Mobile Sources & Fuels, Modeling
Under Agreement No. A932-185, Sierra Research, Inc. (Sierra) has performed several tasks for the Air Resources Board (ARB) related to the characterization of driving patterns and emissions from passenger cars and light trucks operating in California. The major tasks included:
1. Developing of a personal computer version of a vehicle emissions simulation model (named VEHSIME) that uses engine maps and other specified vehicle characteristics to estimate exhaust emissions over any user-specified driving cycle;
2. Using a transportation system planning model to select representative driving routes in the South Coast Air Basin;
3. Using an instrumented "chase car" to record the speed-time profiles of randomly selected vehicles driving over representative driving routes;
4. Constructing a new driving cycle using data collected by the chase car; and
5. Using the emissions simulation model to estimate the difference between emissions on the new cycle and emissions on the "LA4" cycle currently used during the testing of light-duty vehicles under the Federal Test Procedure (FTP).
During the course of the contract, the chase car, equipped with a forward-looking laser range finder system mounted in the front grill, was used to record the speed-time profiles of hundreds of different vehicles driving along a randomly selected sample of road routes in the Greater Metropolitan Los Angeles area. Supplemental information regarding the operation of vehicles at the beginning and end of trips was developed through field surveys.
Analysis of the data indicates that the speed-time profile used in the Federal Test Procedure for light-duty vehicles does not represent the full range of vehicle operation occurring in Los Angeles. Compared to the FTP, the data collected under the contract contain higher acceleration rates and higher speeds.
A methodology was developed for constructing a representative driving cycle by selecting a subset of the available speed-time profiles that closely matches the distribution of velocity and acceleration in the full data set. Using this cycle development methodology, a speed-time profile of approximately the same length as the LA4 was constructed to represent all of the data collected during 1992 using the chase car. As shown in Figure 1, the new "LA92" cycle contains periods of significantly higher vehicle speed than occur on the LA4. The top speed of the new cycle is 108.1 km/hr (67 mph), about 16 km/hr (10 mph) faster than the 91.5 km/hr (56.7 mph) top speed on the LA4. The peak acceleration rate on the new cycle is 3.02 m/s2, over twice as high as the peak acceleration rate of 1.48 m/s2 on the LA4.
Using the VEHSIME computer model to compare exhaust emissions from a typical vehicle driven over the new cycle to one driven over the LA4, CO and NOx emissions are projected to be over twice as high on the new cycle than on the FTP, while HC emissions are nearly the same on both cycles.
Based on the results of the project, the use of a new cycle that better represents current day driving patterns during testing of vehicles recruited from customer service would improve the quality of emissions inventories. In addition, the use of a more representative cycle during certification testing, either as a replacement for or as a supplement to the LA4, would provide an incentive for vehicle manufacturers to design emissions control systems to remain effective during modes of operation that are not currently represented in the official test procedures used during the certification process. However, standards used with a more representative test procedure would have to be different from those used with the LA4 if equivalent stringency is to be maintained. Even uncontrolled vehicles would be expected to emit more CO and NOx emissions on a cycle that includes higher acceleration rates and higher speeds.
One means of adjusting LA4-based standards would be to use the VEHSIME model to estimate emissions on both the LA4 cycle and a new cycle using engine maps that represent uncontrolled engines. Another approach would be to test a representative sample of vehicles without emissions control systems using both cycles.* Either of these approaches could be used to develop adjustment factors that would be applied to LA4-based standards. Alternatively, new standards could be developed based on the results of testing vehicles designed to effectively control emissions during a broader range of operating conditions than are encountered on the LA4 cycle.
The "LA92" cycle developed during the course of the project is clearly superior to the LA4 cycle for representing current day travel in the South Coast Air Basin. Whether ARB should use the LA92 cycle to replace or supplement the LA4 depends on the extent to which consistency with the federal test procedures is to be maintained. Sierra is currently involved in a cycle development effort for EPA which will utilize data collected from instrumented vehicles operating in Baltimore, Maryland. The Baltimore data set is very similar to the LA92 data set in terms of the distribution of vehicle operation by speed and acceleration overall. If a similar cycle is developed for EPA, APB may want to use it for consistency. Alternatively, ARB may want to consider a more comprehensive data collection effort in California.
Before a new cycle is used on a routine basis, further consideration should be given to the treatment of "starts". Under the current test procedure, the LA4 is split into two segments. The first 505 seconds of the cycle are run once using a cold start and once after the vehicle is thoroughly warmed up. One "cold" start and one "hot" start may not be sufficient to adequately represent vehicle emissions during the broad range of conditions that exist at the beginning of trips. In addition, the weighting factors used in the current test procedure may need to be modified to represent the proper proportion of cold and hot start trips. This issue is being addressed under the EPA cycle development effort and the product of that work should be available to ARB in the immediate future.
*Exhaust emission standards based on the LA4 driving cycle are intended to represent a certain degree of control over vehicles with little or no emissions control. For example, the 3.4 g/mi CO standard represents a 90% reduction from the emissions of 1970 model-year federal vehicles, which had relatively little control. The 0.4 g/mi NOx standard represent a 90% reduction from the emissions of 1971 model federal vehicles, which had no NOx control. To the extent emissions only under speeds and loads that occur on the LA4 cycle, an emissions control system design that achieves 90% control over a cycle that is more representative of vehicle operation in customer service. The difference between the emissions of a vehicle on the LA4 cycle and a more representative cycle may not reflect the relative severity of the two cycles unless the vehicle being tested is "uncontrolled."
For questions regarding this research project, including available data and progress status, contact: Heather Choi at (916) 322-3893
Stay involved, sign up with ARB's Research Email Listserver