Model-reference adaptive A/F-ratio control in an SI engine based on Kalman-filtering techniques

Abstract

In this article a model-reference adaptive air/fuel-ratio (A/F-ratio) control technique for SI engines is investigated. The technique is based on the principle of indirect model-reference adaptive control and involves explicit system parameter identification. Since the relevant system dynamics exhibit nonlinearities an extended Kalman-filter is used for parameter identification purposes. The filter is applied online and persistency of excitation is established by introducing supplementary perturbation signals. The resulting parameter estimates are used to update both the controller parameters and the parameters of a state observer providing the nonmeasurable amount of fuel trapped in the cylinders after each intake stroke. This quantity is used for feedback yielding increased control bandwidth. In a series of experimental tests the proposed controller proves to be a highly efficient tool to compensate for the wall-wetting dynamics.