Abstract
For avoiding knock occurrence in SI engines, spark timing is retarded whenever the knock has occurred which leads to a loss of thermal efficiency. Therefore, the knock occurrence needs to be properly controlled. For doing that, knock should preemptively be predicted and controlled. Prerequisite data for knock prediction modelling is a knock onset position, which can be figured out by finding the starting point of the oscillation on pressure data. A deep learning knock onset determination model was developed in a previous study, and showed the highest accuracy among the comparable methods, the model showed weak robustness on knock cycles obtained in different engine experiments. Meanwhile, the 1D CNN model has been widely used in signal processing fields with its advantage of having a feature extraction layer, and the model is introduced in this study for determining the knock onset. Dataset from four different engine types were used for verifying the model accuracy and robustness. The dataset was augmented by calculation windows for producing various data with limited data sources. Hyperparameters of the model were optimized with random search. The accuracy standard deviation following engine types in terms of RMSE was improved by 77.4 % from 0.827 CA to 0.187 CA.
Similar content being viewed by others
References
Arce, G. and McLoughlin, M. (1987). Theoretical analysis of the max/median filter. IEEE Trans. Acoustics, Speech, and Signal Processing 35, 1, 60–69.
Bergstra, J. and Bengio, Y. (2012). Random search for hyper-parameter optimization. J. Machine Learning Research 13, 2, 281–305.
Burluka, A. A., Liu, K., Sheppard, C. G. W., Smallbone, A. J. and Woolley, R. (2004). The influence of simulated residual and NO concentrations on knock onset for PRFs and gasolines. SAE Trans., 1873–1889.
Caruana, R., Lawrence, S. and Giles, C. (2000). Overfitting in neural nets: Backpropagation, conjugate gradient, and early stopping. Proc. 13th Int. Conf. Advances in Neural Information Processing Systems (NIPS), Denver, Colorado, USA.
Cho, S. (2018). Study on the Effect of Cylinder Wall Temperatures on Knock Characteristics in Spark-Ignited Engine. Ph. D. Dissertation. Seoul National University. Seoul, Korea.
Cho, S., Oh, S., Song, C., Shin, W., Song, S., Song, H. H., Min, K., Lee, B., Jung, D. and Woo, S. H. (2019a). Effects of bore-to-stroke ratio on the efficiency and knock characteristics in a single-cylinder GDI engine. SAE Paper No. 2019-01-1138.
Cho, S., Park, J., Song, C., Oh, S., Lee, S., Kim, M. and Min, K. (2019b). Prediction modeling and analysis of knocking combustion using an improved 0D RGF model and supervised deep learning. Energies 12, 5, 844.
Cho, S., Song, C., Kim, N., Oh, S., Han, D. and Min, K. (2021). Influence of the wall temperatures of the combustion chamber and intake ports on the charge temperature and knock characteristics in a spark-ignited engine. Applied Thermal Engineering, 182, 116000.
d’Adamo, A., Breda, S., Fontanesi, S., Irimescu, A., Merola, S. S. and Tornatore, C. (2017). A RANS knock model to predict the statistical occurrence of engine knock. Applied Energy, 191, 251–263.
Dougherty, G. (2012). Pattern Recognition and Classification: An Introduction. Springer Science & Business Media. New York, NY, USA.
Dues, S. M., Adams, J. M. and Shinkle, G. A. (1990). Combustion knock sensing: Sensor selection and application issues. SAE Trans., 1192–1202.
Gao, X., Stone, R., Hudson, C. and Bradbury, I. (1993). The detection and quantification of knock in spark ignition engines. SAE Paper No. 932759.
Guiñón, J. L., Ortega, E., García-Antón, J. and Pérez-Herranz, V. (2007). Moving average and Savitzki-Golay smoothing filters using Mathcad. Int. Conf. Engineering Education (ICEE), Coimbra, Portugal.
Hawkins, D. M. (2004). The problem of overfitting. J. Chemical Information and Computer Sciences 44, 1, 1–12.
Heywood, J. B. (1988). Internal Combustion Engine Fundamentals (1st edn). McGraw-Hill Education. New Yortk, NY, USA.
Kim, K. S. (2015). Study of Engine Knock Using a Monte Carlo Method. Ph. D. Dissertation. The University of Wisconsin-Madison. Madison, Wisconsin, USA.
Kingma, D. P. and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv: 1412. 6980.
Kirsch, L. J. and Quinn, C. P. (1977). A fundamentally based model of knock in the gasoline engine. Symp. (Int.) on Combustion 16, 1, 233–244.
Kotsiantis, S. B., Kanellopoulos, D. and Pintelas, P. E. (2006). Data preprocessing for supervised leaning. Int. J. Computer Science 1, 2, 111–117.
Krizhevsky, A., Sutskever, I. and Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Proc. 25th Int. Conf. Advances in Neural Information Processing Systems (NIPS), Lake Tahoe, Nevada, USA.
LeCun, Y., Bengio, Y. and Hinton, G. (2015). Deep learning. Nature 521, 7553, 436–444.
LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proc. IEEE 86, 11, 2278–2324.
Lee, S., Lee, Y., Lee, Y., Kim, M., Shin, S., Park, J. and Min, K. (2020). EGR prediction of diesel engines in steady-state conditions using deep learning method. Int. J. Automotive Technology 21, 3, 571–578.
Lee, S., Lee, Y., Lee, Y., Shin, S., Kim, M., Park, J. and Min, K. (2021). Proposal of a methodology for designing engine operating variables using predicted NOx emissions based on deep neural networks. J. Mechanical Science and Technology, 35, 1747–1756.
Leppard, W. R. (1982). Individual-cylinder knock occurence and intensity in multicylinder engines. SAE Paper No. 820074.
Livengood, J. C. and Wu, P. C. (1955). Correlation of autoignition phenomena in internal combustion engines and rapid compression machines. Symp. (Int.) on Combustion 5, 1, 347–356.
Nates, R. J. and Yates, A. D. B. (1994). Knock damage mechanisms in spark-ignition engines. SAE Trans., 1970–1980.
Netzer, C., Seidel, L., Pasternak, M., Lehtiniemi, H., Perlman, C., Ravet, F. and Mauss, F. (2018). Three-dimensional computational fluid dynamics engine knock prediction and evaluation based on detailed chemistry and detonation theory. Int. J. Engine Research 19, 1, 33–44.
Oh, S., Cho, S., Shin, W., Min, K. and Song, H. H. (2019). Experimental study on the knock phenomena in the individual cycles of direct-injected spark ignition engine with various stroke-to-bore ratios. 9th European Combustion Meeting (ECM), Lisboa, Portugal.
Park, J., Lee, S., Cho, S., Shin, S., Kim, M., Song, C. and Min, K. (2021). Improvement of knock onset determination based on supervised deep learning using data filtering. SAE Paper No. 2021-01-0383.
Perez, L. and Wang, J. (2017). The effectiveness of data augmentation in image classification using deep learning. arXiv: 1712. 04621.
Real, E., Aggarwal, A., Huang, Y. and Le, Q. V. (2019). Regularized evolution for image classifier architecture search. Proc. 33rd AAAI Conf. Artificial Intelligence, Honolulu, Hawaii, USA.
Redmon, J., Divvala, S., Girshick, R. and Farhadi, A. (2016). You only look once: Unified, real-time object detection. Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Las Vegas, Nevada, USA.
Saha, S. (2018). A comprehensive guide to convolutional neural networks—The ELI5 way. Towards Data Science, 15, 15.
Savitzky, A. and Golay, M. J. (1964). Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry 36, 8, 1627–1639.
Shahlari, A. J. and Ghandhi, J. B. (2012). A comparison of engine knock metrics. SAE Paper No. 2012-32-0007.
Shin, S., Lee, S., Kim, M., Park, J. and Min, K. (2020a). Deep learning procedure for knock, performance and emission prediction at steady-state condition of a gasoline engine. Proc. Institution of Mechanical Engineers, Part D: J. Automobile Engineering 234, 14, 3347–3361.
Shin, S., Lee, Y., Kim, M., Park, J., Lee, S. and Min, K. (2020b). Deep neural network model with Bayesian hyperparameter optimization for prediction of NOx at transient conditions in a diesel engine. Engineering Applications of Artificial Intelligence, 94, 103761.
Young, S. R., Rose, D. C., Karnowski, T. P., Lim, S. H. and Patton, R. M. (2015). Optimizing deep learning hyper-parameters through an evolutionary algorithm. Proc. Workshop on Machine Learning in High-Performance Computing Environments (MLHPC), Austin, Texas, USA.
Yu, D., Wang, H., Chen, P. and Wei, Z. (2014). Mixed pooling for convolutional neural networks. 9th Int. Conf. Rough Sets and Knowledge Technology (RSKT), Shanghai, China.
Zhang, W., Peng, G., Li, C., Chen, Y. and Zhang, Z. (2017). A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors 17, 2, 425.
Zhen, X., Wang, Y., Xu, S., Zhu, Y., Tao, C., Xu, T. and Song, M. (2012). The engine knock analysis–An overview. Applied Energy, 92, 628–636.
Acknowledgement
This research was the results of a study on the “HPC Support” Project, supported by the ‘Ministry of Science and ICT’ and NIPA. The test facility was supported by the Advanced Automotive Research Center (AARC) and the Institute of Advanced Machinery and Design (IAMD) of Seoul National University (SNU). The authors would like to express our deep gratitude to Professor Jo Seok-won from Mississippi State University for his significant help and support in acquiring engine experimental data and writing this paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Park, J., Shin, S., Oh, S. et al. Knock Onset Determination with 1D CNN Using Random Search Hyperparameter Optimization and Data Augmentation in SI Engine. Int.J Automot. Technol. 24, 1395–1410 (2023). https://doi.org/10.1007/s12239-023-0113-7
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12239-023-0113-7