A menthod for predicting in—cylinder compound combustion emissions

This paper presents method using a large steady-state engine operation data matrix to provide necessary information for successfully training a predictive network,while at the same time eliminating errors produced by the dispersive effects of the emissions measurement system.The steady-state training conditions of compund fuel allow for the correlation of time-averaged in -cylinder combustion variables to the engine-out Nox and HC emissions.The error back-propagation neural network(EBP)is then capable of learming the relationships between these variables and the measured gaseous emissions,and then interpolating between steadystate points in the matrix.This method for NOx and HC has been proved highly successful.