Multi-task engine knock detection and intensity estimation based on temporal convolutional networks
Measurement Science and Technology · 24.07.2026 · 37 (29) · 296103
DOI: 10.1088/1361-6501/ae892b ↗Kurz gefasst
Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 21.07.2026. Verfasst von Pengyuan Sun und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 33 Literaturangaben. Gefördert von: Jilin Provincial Scientific and Technological Development Program.
Abstract (Originalsprache)
To address the challenge of reliable engine knock detection under diverse operating conditions, this paper proposes a multi-task framework based on a multi-head temporal convolutional network (TCN) for simultaneous detection and intensity estimation. In-cylinder pressure signals are transformed into a multi-channel representation through Savitzky–Golay detrending and engine-speed-adaptive angle-domain bandpass filtering to suppress low-frequency combustion trends and enhance oscillatory knock-related components. The proposed framework employs a multi-head TCN to regress knock intensity, quantified by the root mean square of the bandpass-filtered residual within a spark-relative or adaptive sliding window, and to classify knock occurrence, enabling a comprehensive evaluation of combustion stability. An operating-condition-adaptive multi-feature labeling strategy with robust normalization and Otsu-guided thresholding is adopted to ensure consistent supervision across varying engine states. To alleviate the adverse effects of class imbalance between knock and non-knock cycles, weighted sampling and class-balanced loss functions are incorporated to improve training stability and generalization. Experimental evaluations across multiple engine speeds and loads demonstrate that the proposed method surpasses conventional threshold-based indices and recent deep learning baselines, achieving superior accuracy, robustness, and computational efficiency suitable for real-time combustion monitoring and control-oriented applications. To ensure the reliability of the reported results, all metrics were averaged over three independent runs to mitigate the impact of network initialization and sampling randomness.
Förderung
- Jilin Provincial Scientific and Technological Development Program
Zitieren
APA
Sun, P., Zhao, H., Qu, H., Gao, T., Chen, W., & Xu, C. (2026). Multi-task engine knock detection and intensity estimation based on temporal convolutional networks. Measurement Science and Technology, 37(29), 296103. https://doi.org/10.1088/1361-6501/ae892b
Vancouver
Sun P, Zhao H, Qu H, Gao T, Chen W, Xu C. Multi-task engine knock detection and intensity estimation based on temporal convolutional networks. Measurement Science and Technology. 2026;37(29):296103. doi:10.1088/1361-6501/ae892b
BibTeX
@article{Sun2026,
author = {Sun, Pengyuan and Zhao, Huichao and Qu, Hanshi and Gao, Tianyu and Chen, Weixuan and Xu, Chengcheng},
title = {Multi-task engine knock detection and intensity estimation based on temporal convolutional networks},
journal = {Measurement Science and Technology},
year = {2026},
volume = {37},
number = {29},
pages = {296103},
doi = {10.1088/1361-6501/ae892b}
}
EUVIMED