Misfire Detection in a High-Performance Engine by the Principal Component Analysis Approach

Format:

Conference/Event

Author/Creator:

Azzoni, Piero Mario, author.

Conference Name:

International Congress & Exposition (1996-02-26 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 1996

Summary:

The aim of this paper is to present the application of some signal processing and statistical analysis methods to the problem of detecting and isolating misfire occurrences in a twelve-cylinder high-performance engine. The method employed in this work is based on a measurement of engine angular velocity, processed in the frequency domain to extract a number of spectral components that are shown to be strongly affected by misfire events. These spectral components are then subject to a procedure known as Principal Components Analysis, in which the principal features of the angular speed waveform are extracted to generate individual cylinder misfire signatures. A clustering method is then implemented to permit the isolation of the cylinder responsible for the misfire. The paper briefly reviews the signal analysis method and presents experimental results supporting the validity of the approach

Notes:

Vendor supplied data

Publisher Number:

960622

Access Restriction:

Restricted for use by site license