The research objective of this project is to increase the understanding of engine knocking and find novel methods to develop ultra-fast knock detection algorithms that are able to detect knocking in real time. Moreover, a step towards predictive knock control is taken, where the algorithm tries to predict knocking and takes measures to adjust the ignition conditions in the combustion cycle even before knocking occurs. Since the common physics-based models are already comparatively well developed and have undergone extensive investigations in the past, the novel methods aimed at in this project will primarily leverage data analytics and machine learning used for pattern recognition. In combination with the classic physical methods in hybrid approaches, they have the potential to improve knock prediction.
|Effective start/end date
|1/01/20 → 31/12/23
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.