A Portable Emission Measurement System (PEMS) was utilised to record the on-road Carbon Dioxide (CO2) emission of a EURO 4 petrol vehicle over 48 test runs through an urban-traffic network. The tests were conducted over a 780 metre micro-scale road segment between Headingley and the City of Leeds, UK, with measurement on both the inbound (Section A) and outbound lanes (Section B). The monitored test runs were conducted under a range of traffic flow conditions from heavily congested to free-flowing traffic. Vehicle exhaust emission simulations using an instantaneous power-emission model have the capability to generate estimates of real-world vehicle emissions over micro-scale road sections. The Technical University of Graz’s (TUG) Passenger car and Heavy duty Emission Model (PHEM) was used to calculate a CO2 emission estimate for each of the 48 test runs through Sections A and B. The model CO2 emission estimates were then compared to the real-world PEMS emission measurements, to determine the accuracy of the modelling methodology. Whilst instrumented vehicles can adequately capture second-by-second (1Hz) absolute position and vehicle speed there is significant instrument error in the measurement of real-world elevation using a Global Positioning System (GPS) as part of a PEMS set-up. These errors make it very difficult to accurately calculate a 1Hz road grade with GPS systems. However, as road grade can have an important influence on engine power demand and hence fuel consumption and exhaust emission it is essential to include a representative road grade estimate for micro-scale emission estimation. Rather than using a GPS recorded elevation, this study developed a simple road grade estimation methodology which employs Geographic Information System (GIS) software to interpolate the elevation at each second of PEMS data from a 5-metre resolution Digital Terrain Map (DTM) derived from Light Detection And Ranging (LiDAR) data. The method applies an algorithm to compute the road grade from the LiDAR-GIS elevation values and vehicle speed, and alleviates errors resulting from absolute position measurement inaccuracy of the GPS at low speed. The addition of the LiDAR-GIS road grade to the PHEM modelling was found to improve the accuracy of the PHEM estimate of the PEMS measured real-world CO2 emission. From the 48 test runs the average PHEM estimate (including road grade) of the real-world measured CO2 emission through Section A was 93%, and through Section B was 94%. Of the total 96 test runs over Section A and B 91% of the PHEM estimates were between 80% and 110% of the PEMS recorded value. In further analysis, an assessment of the effect of road grade on both CO2 and NOx emission was conducted. Sections A and B were combined for each test run to form Segment AB, which has a net flat road grade. The PEMS recorded speed profiles for each of the test runs through sections A and B were input into PHEM and emission estimates generated under four road grade scenarios. The scenarios were formed by decreasing and exaggerating the LiDAR-GIS road grade for each second of data, multiplying it by coefficients of 0 (flat), 0.5 (half the grade), 1, and 2 (double the grade). The results indicate that assuming a flat profile in PHEM would result in an average underestimate of the segment emission by 2.7% for CO2 and 7.0% for NOx when calculated with road grade, and by 7.9% for CO2 and 20.4% for NOx were the road grade doubled. The method developed in this study provides a simple methodology for calculating 1Hz road grade, and has been shown to improve the modelling of CO2 emission for this data set. This research suggest that using the PHEM model with a LiDAR-GIS calculated road grade provides a practical method for accurately estimating real-world micro-scale emission. On-road emission monitoring by PEMS is scheduled to be introduced for Euro 6c type approval from September 2017. In order to accurately determine road load during the real-world test procedure it will be important to develop a suitable methodology for calculating a 1Hz road grade.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)