Core Concepts
Data-based approach for in-cylinder pressure modeling with cycle-to-cycle variations for RCCI engines.
Abstract
Cylinder pressure-based control is crucial for advanced combustion concepts.
Fast control-oriented combustion models are needed for robust and safe operation.
Data-based approach models in-cylinder pressure and cycle-to-cycle variations.
Principle Component Decomposition and Gaussian Process Regression are used.
Model applicable to combustion concepts with large cycle-to-cycle variation.
Model demonstrates high accuracy in predicting combustion measures.
Importance of in-cylinder pressure shaping for advanced combustion concepts.
Gaussian Process Regression provides valuable information for control design.
Study includes detailed analysis of hyperparameters and kernel choices.
Model shows potential for cycle-to-cycle variation control and safety criteria.
Stats
Die Genauigkeit der vorhergesagten Verbrennungsmessungen beträgt insgesamt 13,5% und 65,5% in Bezug auf das mittlere Verhalten und die Standardabweichung.
Die Genauigkeit des Anstiegs des Spitzeninnendrucks beträgt 22,7% und 96,4% in Bezug auf das mittlere Verhalten und die Standardabweichung.
Quotes
"Cylinder pressure-based control is a key enabler for advanced pre-mixed combustion concepts."
"The model combines Principle Component Decomposition and Gaussian Process Regression."
"The approach is applicable to any combustion concept, but most valuable for advance combustion concepts with large cycle-to-cycle variation."