KNOCK ONSET DETERMINATION WITH 1D CNN USING RANDOM
SEARCH HYPERPARAMETER OPTIMIZATION AND DATA
AUGMENTATION IN SI ENGINE

, Jihwan Park; , Seunghyup Shin; , Sechul Oh; , Sangyul Lee; , Woojae Shin; , Kyoungdoug Min

doi:2023.24.5.1395

International Journal of Automotive Technology > Volume 24(5); 2023 > Article

Electric, Fuel Cell, and Hybrid Vehicle, Engine and Emissions, Fuels and Lubricants, Heat Transfer, Fluid and Thermal Engineering

International Journal of Automotive Technology 2023;24(5): 1395-1410.
doi: https://doi.org/10.1007/s12239-023-0113-7

KNOCK ONSET DETERMINATION WITH 1D CNN USING RANDOM SEARCH HYPERPARAMETER OPTIMIZATION AND DATA AUGMENTATION IN SI ENGINE

Jihwan Park ¹, Seunghyup Shin ², Sechul Oh ³, Sangyul Lee ⁴, Woojae Shin ¹, Kyoungdoug Min ¹

¹Department of Mechanical Engineering, Seoul National University
²Department of Artificial Intelligence, Sejong University
³Department of Mobility Power Research, Eco-Friendly Energy Conversion Research Division, Korea Institute of Machinery and Materials
⁴Department of Mechanical and Electronics Engineering, Hansung University

Corresponding Author. Kyoungdoug Min , Email. kdmin@snu.ac.kr

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.

Key Words: Spark ignition engine, Knock, Knock onset determination, Deep learning, 1D convolution neural network, Data augmentation