Exploring the Accuracy of Machine Learning and Deep Learning in Engine Knock Detection

Authors

  • Usman Hameed Department of Computer Science, Superior University, Lahore, 54000, Pakistan Author
  • Dr. Sohail Masood Professor, Department of Computer Science, Superior University, Lahore, 54000, Pakistan Author
  • Fawad Nasim Department of Computer Science, Superior University, Lahore, 54000, Pakistan Author
  • Dr. Arfan Jaffar Department of Computer Science, Superior University, Lahore, 54000, Pakistan Author

DOI:

https://doi.org/10.61506/01.00478

Keywords:

Machine Learning, Engine knocking, FMAD, Deep learning.

Abstract

This study explores the use of machine learning for real-time detection of engine knocking, aiming to enhance early vehicle fault recognition. We extracted frequency modulation amplitude demodulation (FMAD) features from engine sound data and evaluated various machine-learning algorithms using MATLAB. The coarse decision tree algorithm emerged as the most effective, achieving a classification accuracy of 66.01%. Subsequently, by using deep learning models, we significantly improved the accuracy: a convolutional neural network (CNN) achieved 45.16%.  accuracy, a deep learning recurrent neural network (RNN) model in LSTM achieved 90% accuracy, and further refinements pushed the accuracy to 93.55%. Additionally, we introduced a knock index to quantify noise levels during each engine cycle. This index, calculated from the integral of the absolute value of the first derivative of a band-pass-filtered vibration signal, provides a visual representation of knock strength. This approach shows promise for early detection of engine knocking, although further refinement of feature extraction methods and algorithm optimization is necessary for practical application. The study highlights the potential of integrating machine learning into real-time vehicle fault detection systems to improve their reliability and effectiveness.

References

Aramburu, A., Guido, C., Bares, P., Pla, B., Napolitano, P., & Beatrice, C. (2024). Knock detection in spark ignited heavy-duty engines: An application of machine learning techniques with various knock sensor locations. Measurement, 224, 113860. DOI: https://doi.org/10.1016/j.measurement.2023.113860

Hosseini, M., & Chitsaz, I. (2023). Knock probability determination employing convolutional neural network and IGTD algorithm. Energy, 284, 129282. DOI: https://doi.org/10.1016/j.energy.2023.129282

Hosseini, M., & Chitsaz, I. (2023). Knock probability determination in a turbocharged gasoline engine through exhaust gas temperature and artificial neural network. Applied Thermal Engineering, 225, 120217. DOI: https://doi.org/10.1016/j.applthermaleng.2023.120217

Kalghatgi, G. T., & Bradley, D. (2012). Pre-ignition and ‘super-knock’in turbo-charged spark-ignition engines. International Journal of Engine Research, 13(4), 399-414. DOI: https://doi.org/10.1177/1468087411431890

Hosogi, T., Kaneko, R., Ito, H., & Sunami, K. (2019). Research of Knocking Deterioration due to Accumulated Carbon Deposits on Piston Surfaces (No. 2019-01-1141). SAE Technical Paper. DOI: https://doi.org/10.4271/2019-01-1141

Sun, J., Zhang, X., Tang, Q., Wang, Y., & Li, Y. (2023). Knock recognition of knock sensor signal based on wavelet transform and variational mode decomposition algorithm. Energy Conversion and Management, 287, 117062. DOI: https://doi.org/10.1016/j.enconman.2023.117062

Meier, R., Pacheco, A. F., Cancino, L. R., & Oliveira, A. A. (2024). Effect of first-stage ignition on the onset of knocking in SI engines. International Journal of Engine Research, 25(1), 65-84. DOI: https://doi.org/10.1177/14680874231193207

Meng, S., Han, Z., Fan, B., Wu, Z., Shao, Q., & Tong, L. (2024). Experimental study of spatial distribution of knock events in a turbocharged spark-ignition engine. International Journal of Engine Research, 25(7), 1251-1264. DOI: https://doi.org/10.1177/14680874241227552

Sann, S., & Tomeh, E. (2024). In-cylinder pressure and vibration analysis on port fuel injected gasoline engine. Vibroengineering Procedia, 54, 15-21. DOI: https://doi.org/10.21595/vp.2024.23960

Mittal, V. (2024). A Review of Recent Advancements in Knock Detection in Spark Ignition Engines. Signals, 5(1), 165-180. DOI: https://doi.org/10.3390/signals5010009

Suijs, W., Dierickx, J., Pu, Y. H., Wang, Y., & Verhelst, S. (2024). Calibrating the Livengood–Wu integral knock model for differently sized methanol engines. Fuel Communications, 19, 100121. DOI: https://doi.org/10.1016/j.jfueco.2024.100121

Suijs, W., Broekaert, S., De Cuyper, T., & Verhelst, S. (2024). The sensitivity of pressure-based knock threshold values to alternative fuels: A comparison of methanol vs. gasoline. Fuel, 362, 130850. DOI: https://doi.org/10.1016/j.fuel.2023.130850

Nasim, F., Masood, S., Jaffar, A., Ahmad, U., & Rashid, M. (2023). Intelligent Sound-Based Early Fault Detection System for Vehicles. Computer Systems Science & Engineering, 46(3). DOI: https://doi.org/10.32604/csse.2023.034550

Downloads

Published

2024-08-28

Issue

Section

Articles

How to Cite

Hameed, U. ., Masood, S. ., Nasim, F. ., & Jaffar, A. . (2024). Exploring the Accuracy of Machine Learning and Deep Learning in Engine Knock Detection. Bulletin of Business and Economics (BBE), 13(3), 203-210. https://doi.org/10.61506/01.00478