Automotive engine misfire detection using Kalman filtering

This paper presents a new approach to misfire detection using system parameter estimation techniques. A mathematical model has been developed for the engine firing system. The resulting model contains many unknown parameters or coefficients. A system parameter identification technique employing a Kalman filter is then developed to estimate all the unknown parameters based on actual vehicle test data. The paper shows that the new Kalman filter approach for misfire detection has greatly enhanced the detection accuracy and reduced the false alarm rate of current misfire detection systems.