Search Results

In this paper we present an advanced knock detection approach. The detection concept consists of a two-level feature extraction step followed by neural network detector. A knock tendency index is estimated that takes into account the statistical behavior of the knock phenomena. The configuration of the neural network is based on a signal database that was acquired under almost ‘on-road’ conditions. The experimental set-up consisted of several measurement sessions in a special vehicle test cell. In order to achieve a most realistic knock database the test engine was mounted on an in-production car.

This article describes an application of sliding mode techniques to the design of an air/fuel ratio control system for a 4-stroke engine, to minimize exhaust gas and emissions. This technique allows to achieve good control performance in terms of precision, robustness, and fast transient response. To support sliding mode control a second PI stage was added, based on the signal of a second oxygen sensor installed after the catalytic converter. Experimental results were better than those obtained with a conventional PI control, currently used on production applications. The new control algorithm (sliding mode based on the first oxygen sensor, and PI on the second) is very versatile because the approach chosen allows to calculate the parameters values for the ECU using computer simulation.