Uncertainties in Model-Based Diesel Particulate Filter Diagnostics Using a Soot Sensor

Abstract

Monitoring the filtration efficiency of the diesel particulate filter (DPF), is a legislative requirement for minimizing particulate matter (PM) emissions from diesel engines of passenger cars and heavy-duty vehicles. To reach this target, on-board diagnostics (OBD) in real-time operation are required. Such systems in passenger cars are often utilizing a soot sensor, models for PM emissions simulation and algorithms for diagnosis. Their performance is associated with a series of challenges related to the accuracy and effectiveness of involved models, algorithms and hardware. This paper analyzes the main influencing factors and their impact on the effectiveness of the OBD system. The followed method comprised an error propagation analysis to quantify the error of detection during a New European Driving Cycle (NEDC). The results of the study regarding the performance of the OBD model showed that the total error of diagnosis is ±28%. This performance can be improved by increasing the sensor accuracy and the soot model, which can make the model appropriate for even tighter legislation limits and other approaches such as on-board monitoring (OBM).