Engine Control Optimization Via Nonlinear Programming

A procedure has been developed for establishing optimum engine performance, in terms of fuel economy and exhaust emission constraints, as a function of spark advance, exhaust gas recirculation and air-to-fuel ratio. The salient features of the procedure are: (i) it is based on mathematical engine models that have been validated over a driving cycle on an engine dynamometer, (ii) it is based on solving a nonlinear programming problem with equality and inequality constraints, and (iii) the results of the optimization are used to develop control calibrations that can be implemented on a vehicle. This paper describes the procedure to determine optimal calibrations for a fuel injected spark ignition engine under warmed-up operating conditions. Validation of these results on an actual vehicle has yet to be demonstrated.