Abstract
The use of a continuously variable transmission (CVT) with an integrated engine controller allows great freedom of choice of the engine operating point to deliver the power demanded by the driver. This flexibility places even greater emphasis on the requirement to locate an appropriate engine operating point. A technique has been developed that locates the ideal instantaneous engine operating point by minimizing a simple weighted sum of exhaust emissions and fuel consumption predicted by an artificial neural network. This process may be configured automatically to take account of varying operating or environmental conditions, allowing the optimum performance to be returned by the powertrain controller at all times. The structure has been implemented for a turbocharged and intercooled direct injection diesel engine used as the prime mover in an integrated powertrain. Results are presented that demonstrate the use of the optimizer to derive widely varying operating schedules for the engine, each with associated emissions and economy implications. The optimizer code is sufficiently compact to be incorporated into a practical powertrain control strategy. Experimental work presented in a companion paper has confirmed that this 'systems' approach to powertrain control allows the emissions and economy performance of the vehicle to be tailored to suit the particular requirements of the installation.
Original language | English |
---|---|
Pages (from-to) | 215-226 |
Number of pages | 12 |
Journal | Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering |
Volume | 213 |
Issue number | 3 |
DOIs | |
Publication status | Published - 1 Mar 1999 |
Keywords
- Continuously variable transmissions
- Diesel engines
- Emissions
- Fuel economy
- Integrated powertrain control
- Neural networks
- Powertrain calibration
ASJC Scopus subject areas
- Aerospace Engineering
- Mechanical Engineering