Robust combined feedforward and feedback control for start up engine control

In this paper, a robust combined feedforward and feedback control for start-up engine control for initial state uncertainties and disturbances is proposed. In the proposed control method, the feedforward control inputs are searched by a numerical optimization method called PSO (particle swarm optimization) for extremum initial conditions and they are generalized by an ANN (artificial neural network) to cope with the uncertainties. The feedback control is realized by GPC (generalized predictive control) based on JIT (just in time) modeling of a local plant model. The validity of the proposed method is demonstrated by numerical simulations.