Acoustic Radiation of a Diesel Engine Under Air Flow Meter Fault

This chapter delves into the critical role of acoustic analysis in detecting faults in diesel engine airflow sensors. The study presents a comprehensive methodology for collecting and analyzing acoustic data, utilizing advanced statistical indicators such as kurtosis, standard deviation, and root mean square (RMS) to identify faults. The research demonstrates how these indicators can reliably detect early signs of sensor malfunctions, particularly at higher engine speeds. The experimental setup involves a detailed description of the acoustic data collection system, including the use of a Class 2 sound level meter compliant with IEC 61672–1. The results highlight significant deviations in acoustic signals under faulty conditions, providing valuable insights into combustion anomalies. The study concludes by emphasizing the potential of acoustic-based monitoring for early fault detection, offering a cost-effective and non-intrusive method to enhance engine diagnostics, reduce maintenance costs, and improve overall system reliability. Future research directions include expanding this methodology to other engine components and integrating machine learning techniques for automated fault identification.