Methodology for Detecting Faults in Injection Systems of Low-Displacement Diesel Engines by Diagnosing the Vibration Signal

Abstract

In engines, the injection process is directly related to the atomization control and the detonation of the combustion mixture. A malfunction in this process caused by injectors’ failures leads to reduced engine efficiency, excessive noise, and damage. In this study, an analysis is proposed in order to detect problems in the injector by diagnosing the vibration signal of the engine body, especially the amplitude and the frequency of this type of signal. In order to carry out this study, a cylindrical mono diesel engine that varies its rotation speed by 3200, 3600, and 4000 rpm, at a constant torque of 3 Nm is used. For the vibration signal analysis, the study of the signal frequency, the Welch test, and the Fourier transform in the short term has been used. Additionally, parameters such as RMS and Kurtosis, which allow quantifying the effect of the injector on engine performance, have been evaluated. The analysis of the low and high frequency diagrams has proved to be a direct way to identify the status of the injectors. The Welch test allows observing that the defective injectors have a vibration signal intensity 44% higher than the healthy injectors. In general, the use of defective injectors has caused an 11% and 17% increase in the RMS and Kurtosis of the engine, compared to healthy injectors. In general, the study carried out shows that the fault detection in the injectors of IC engines is possible by measuring and evaluating the time-frequency of vibration signals from the engine body.