Assessing the influence of boundary conditions, driving behavior and data analysis methods on real driving CO and NO emissions.

The need to verify vehicle emissions in real world operation led to the implementation of Real Driving Emissions (RDE) test procedures, effective since September 2017 for new Euro 6 cars following the Commission Regulation (EU) 2016/427, which defines the RDE test conditions and data analysis methods to allow representative results. Main factors addressed by the regulation include the share of driving operation, ambient temperature range, altitude and elevation difference. However, RDE is still debatable since not only boundary conditions but also the evaluation methods and trip selection are being discussed together with a carbon dioxide (CO) regulation, which is planned to be implemented in the short term. Thus, this work focuses on analyzing the effect of different data measurement and analysis methods (i.e. cold-operation, road grade, trip selection and driving style) on CO and nitrogen oxides (NO) emissions based on 13 RDE tests performed in the Lisbon Metropolitan Area, Portugal. The tests were conducted by 2 drivers using 5 vehicles. Each driver performed 2 trips per vehicle, one in normal driving and other in aggressive driving. A Portable Emissions Measurement System (PEMS) was used to collect 1 Hz data, which was compared and analyzed using the European Commission (EC) proposed method for RDE tests. Results show the effects of each parameter such as average difference between drivers (7% in CO and 55% in NO emissions) and between aggressive and normal driving. For road grade, big oscillations happen during the slope profile, which impacts emissions during all trips. Considering cold-operation, CO and NO emissions are, on average, ~25% and 55% higher, respectively, than in hot-operation. These results highlight the need for deeper studies on these factors to assure that RDE tests evolve to a more established certification procedure than laboratorial certifications.