Real-time Adaptive Predictive Control of the Diesel Engine Air-path Based on Fuzzy Parameters Estimation

In this paper, a robust adaptive optimal tracking control design for the air-path system of diesel engines with uncertain parameters and external driver commands is proposed.

First, an optimal controller based on the analytic solution of a performance index is derived. It achieves tracking of suitable references (corresponding to low emissions and fuel consumption) for both the air-fuel ratio and the fraction of the recirculated exhaust gas. Then, a fuzzy estimation algorithm is used to identify the plant parameters and consequently to adapt the controller online.

The simulated diesel engine is a medium duty Caterpillar 3126B with six cylinders, equipped with a variable geometry turbocharger and an exhaust gas recirculation valve. The proposed controller design is based on the reduced third order mean value model and implemented as a closed-form nonlinear model predictive control law on the full order model. The resulting controllers, with and without adaptation, are assessed through simulations with a SIL architecture using dSpace simulator. The adaptive controller, in particular, exhibits good control performance, with zero tracking offset with also ensuring global stability and tracking of output references.