TY - GEN
T1 - Freevalve: Control and Optimization of Fully Variable Valvetrain-Enabled Combustion Strategies for Steady-State Part Load Performance and Transient Rise Times
AU - Elmagdoub, Abdelrahman W.M.
AU - Carlson, Urban
AU - Halmearo, Mattias
AU - Turner, James W. G.
AU - Brace, Chris
AU - Akehurst, Sam
AU - Zhang, Nic
N1 - KAUST Repository Item: Exported on 2023-06-12
Acknowledgements: The authors wish to thank the directors of Koenigsegg Automotive AB and Freevalve AB for permission to publish this paper. The modelling work undertaken was based upon a validated model originally developed under the “Ultra Boost for Economy” project funded by the UK Technology Strategy Board, now Innovate UK (IUK), grant number BN008E 2010- 2013. The contribution of Lotus Engineering towards producing the first-generation GT-Power model used and updated throughout this paper for comparative reasons is also recognized. This work is supported by a scholarship from the UK’s EPSRC Centre for Doctoral Training in Advanced Automotive Propulsion Systems (AAPS) at the University of Bath, under project code EP/S023364/1.
PY - 2023/4/11
Y1 - 2023/4/11
N2 - In passenger car development, extreme ICE downsizing trends have been observed over the past decade. While this comes with fuel economy benefits, they are often obtained at the expense of Brake Mean Effective Pressure (BMEP) rise time in transient engine response. Through advanced control strategies, the use of Fully Variable Valvetrain (FVVT) technologies has the potential to completely mitigate the associated drivability-penalizing constraints. Adopting a statistical approach, key part load performance engine parameters are analyzed. Design-of-Experiment data is generated using a validated GT-Power model for a Freevalve-converted turbocharged Ultraboost engine. Subsequently, MathWorks' Model Based Calibration (MBC) toolbox is utilized to interpret the data through model fitments using neural network models of optimized architectures. Calibration Generation (CAGE) toolbox is ultimately used to identify best-case look-up tables for the part load steady state performance points based on concluded, case specific, BSFC values. Transient tip-in events are simulated using a step pedal input to full load from the optimized part load points and total rise times are analyzed. For conventional non-FVVT configurations it has been demonstrated that part load cases with higher EGR rates concluded significantly higher T10 (time to 10% of BMEP) values, while T90 (time to 90% of BMEP) and T10-90 (time between 10% and 90% of BMEP) at the tip-in transient were least influenced by residual content. Assuming a Pareto optimal front, this leads to propose that advanced valve control strategies enabled by FVVT technologies, targeting maximum scavenging and optimized EGR rates, are capable of eliminating the potential burden that is turbocharger lag, otherwise sustained in boosted engines as a result of limited cam-based valvetrains, on tip-in transient events from a minimum BSFC steady state part load initial condition.
AB - In passenger car development, extreme ICE downsizing trends have been observed over the past decade. While this comes with fuel economy benefits, they are often obtained at the expense of Brake Mean Effective Pressure (BMEP) rise time in transient engine response. Through advanced control strategies, the use of Fully Variable Valvetrain (FVVT) technologies has the potential to completely mitigate the associated drivability-penalizing constraints. Adopting a statistical approach, key part load performance engine parameters are analyzed. Design-of-Experiment data is generated using a validated GT-Power model for a Freevalve-converted turbocharged Ultraboost engine. Subsequently, MathWorks' Model Based Calibration (MBC) toolbox is utilized to interpret the data through model fitments using neural network models of optimized architectures. Calibration Generation (CAGE) toolbox is ultimately used to identify best-case look-up tables for the part load steady state performance points based on concluded, case specific, BSFC values. Transient tip-in events are simulated using a step pedal input to full load from the optimized part load points and total rise times are analyzed. For conventional non-FVVT configurations it has been demonstrated that part load cases with higher EGR rates concluded significantly higher T10 (time to 10% of BMEP) values, while T90 (time to 90% of BMEP) and T10-90 (time between 10% and 90% of BMEP) at the tip-in transient were least influenced by residual content. Assuming a Pareto optimal front, this leads to propose that advanced valve control strategies enabled by FVVT technologies, targeting maximum scavenging and optimized EGR rates, are capable of eliminating the potential burden that is turbocharger lag, otherwise sustained in boosted engines as a result of limited cam-based valvetrains, on tip-in transient events from a minimum BSFC steady state part load initial condition.
UR - http://hdl.handle.net/10754/692538
UR - https://www.sae.org/content/2023-01-0294
UR - http://www.scopus.com/inward/record.url?scp=85160765800&partnerID=8YFLogxK
U2 - 10.4271/2023-01-0294
DO - 10.4271/2023-01-0294
M3 - Conference contribution
BT - SAE Technical Paper Series
PB - SAE International
ER -