Evaluating on-board sensing-based nitrogen oxides (NOX) emissions from a heavy-duty diesel truck in China

Abstract

Heavy-duty diesel vehicles (HDDVs) are important sources of urban nitrogen oxides (NOX). Understanding real-world NOX emissions from HDDVs is important because it has been shown that NOX emissions could far exceed the desired limits and impact human health. On-vehicle sensors for real-time emission evaluations have become commercially available, and China has adopted an on-board sensing (OBS) pilot program in Beijing. The OBS system utilizes information from the engine data networks which including real-time NOX concentrations from on-road HDDVs. To evaluate the accuracy of OBS monitoring, one of the vehicles in the OBS pilot program was equipped with a portable emissions measurement system (PEMS) and was evaluated over the course of two days during on-road comparison testing. The vehicle was operated under normal conditions and under conditions simulating a tampered vehicle with no urea for the selective catalytic reduction (SCR) system. A comparison between the OBS and PEMS exhibited good NOX correlations on an instantaneous and moving-average basis. Although large OBS-to-PEMS discrepancies were found in a small portion of instantaneous measurements, the 60-s moving averaged NOX concentrations exhibited good agreements across all the test conditions (Pearson's R > 0.95 and mean relative errors from −13% to +22%). During the tampered SCR simulation, the average vehicle emissions increased from 22 g kg fuel−1 to 48 g kg fuel−1, where both the PEMS and OBS quantified consistent trends. The actual NOX emissions, registered by the OBS over fifteen consecutive days of normal vehicle operation, ranged from 3.29 g km−1 to 6.65 g km−1. A comparison with the PEMS suggests the OBS can effectively and accurately identify high-emitting situations for in-use diesel vehicles. Recommendations are provided to improve the implementation of OBS programs in future applications.