Misfire Detection Based on Generalized Force Identification at the Engine Centre of Gravity

Chuanyan Xu,Fengping Cao,Shan Li,Xuyun Qiu

doi:10.1109/access.2019.2947621

Chuanyan Xu, Fengping Cao + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2947621

Copy DOI

Abstract

Misfiring creates a unique pattern attributed to a particular cylinder. When a misfire occurs, the balance of the engine is destroyed, and the generalized force at the centre of gravity (C.G.) of the engine is changed. In this paper, a new misfire detection method is presented based on the identification of the generalized force at the engine centre of gravity. Based on the engine acceleration signals at the mounts, through the use of the discrete spectrum interpolation method, the accurate amplitudes and phases of the acceleration signals are extracted, and then, the generalized force at the centre of gravity is calculated. Through analysing the main harmonic orders of the generalized force, the misfire features are accurately extracted and classified. Both the simulation examples and test bed results prove the effectiveness of the present method in detecting misfire faults in combustion engines.

Highlights

Structural dynamical monitoring and fault diagnosis are of great importance to structural health
THE GENERALIZED FORCE ESTIMATION The generalized force at the gravity centre of an engine is normally calculated based on the engine parameters [24], which are the gas explosion force, reciprocating inertial force, and rotating inertia force
The interpolation method using a Hanning window is applied for the extraction of the frequency, amplitude and phase, and the generalized force at the centre of gravity can be accurately identified according to equation (1)

Summary

INTRODUCTION

Structural dynamical monitoring and fault diagnosis are of great importance to structural health. Post-cylinder factor methods apply features such as the instantaneous crankshaft angular velocity [9], instantaneous angular acceleration [10], crankshaft combustion torque [11], crank angle [12], engine vibration signal [13], exhaust gas pressure [14], exhaust gas temperature [15] and engine noise [16], [17] to detect misfire faults. The diagnosis accuracy of vibration-based misfire detection methods is affected by many factors such as complex engine construction, heavy noise, the temperature and the sensor installation position. The method of misfire detection presented in this paper is based on the estimation of the generalized force at the gravity centre of the engine.

THE GENERALIZED FORCE ESTIMATION

EXPERIMENTAL SETUP

CONCLUSION