Misfire Index Design by Statistical Signal Processing in High Performance Engines

Abstract The present paper describes the application of statistical signal processing methods for the detection and identification of engine misfire in a high performance engine. The new environmental regulation in the Unites Stales requires the misfire detection because of its potential effect to decrease the effectiveness of the three way catalyst, at least resulting the irreversible destruction of the catalyst itself if the misfire presence is systematic. The method is based on compression and clustering of a data vector consisting of twelve Fourier components of the engine angular velocity signal. The paper describes the theory of the method and experimental results, illustrating the applicability of the method and its potential ability to solve a heretofore very challenging problem.