Deep feature learning of in-cylinder flow fields to analyze cycle-to-cycle variations in an SI engine. (November 2021)

• Record Type: Journal Article
• Authors: Dreher, Daniel; Schmidt, Marius; Welch, Cooper; Ourza, Sara; Zündorf, Samuel; Maucher, Johannes; Peters, Steven; Dreizler, Andreas; Böhm, Benjamin; Hanuschkin, Alexander
• Is Part Of: International journal of engine research. Volume 22:Number 11(2021)
• Page Start: 3263
• Page End: 3285
• Publication Date: 2021-11
• Subjects: Deep learning -- machine learning -- feature analysis -- particle image velocimetry -- in-cylinder flow -- cycle-to-cycle variations -- IC engine; Engines -- Periodicals; 629.25
• DOI: 10.1177/1468087420974148
• Languages: English
• ISSNs: 1468-0874

Abstract

Machine learning (ML) models based on a large data set of in-cylinder flow fields of an IC engine obtained by high-speed particle image velocimetry allow the identification of relevant flow structures underlying cycle-to-cycle variations of engine performance. To this end, deep feature learning is employed to train ML models that predict cycles of high and low in-cylinder maximum pressure. Deep convolutional autoencoders are self-supervised-trained to encode flow field features in low dimensional latent space. Without the limitations ascribable to manual feature engineering, ML models based on these learned features are able to classify high energy cycles already from the flow field during late intake and the compression stroke as early as 290 crank angle degrees before top dead center (− 290 ° CA ) with a mean accuracy above chance level. The prediction accuracy from− 290 ° CA to− 10 ° CA is comparable to baseline ML approaches utilizing an extensive set of engineered features. Relevant flow structures in the compression stroke are revealed by feature analysis of ML models and are interpreted using conditional averaged flow quantities. This analysis unveils the importance of the horizontal velocity component of in-cylinder flows in predicting engine performance. Combining deep learning and conventional flow analysis techniques promises to be a powerful tool for ultimately revealing high-level flow features relevant to the prediction of cycle-to-cycle variations and further engine optimization.