Single NLARX Model for Particulate Matters Prediction of Diesel Engines
World Congress, Volume # 18 | Part# 1
Deng, Jiamei; Maass, Bastian; Stobart, Richard
Digital Object Identifier (DOI)
Modeling, supervision, control and diagnosis of automotive systems; Neural networks; Modeling and simulation of transportation systems
TThis paper describes a neural network that can be used as a virtual sensor for measuring particulate matter (PM) emissions of a medium or heavy-duty diesel engine. The neural network is stable across a broad range of engine operation points. The input parameters are chosen based on the PM formation mechanism, physical knowledge of the process and an insight of the underlying physics. The results show that neural network models could predict the particulate matter successfully with R2 above 0.99 with 5 inputs only.
No references available