Automotive engine modelling based on online time-sequence incremental and decremental least-squares support vector machines

摘要

Air-ratio relates closely to engine emissions, power and fuel consumption among all of the engine parameters. The thesis proposed an online time-sequence incremental and decremental least-squares support vector machines (OLSSVM) for engine modelling to predict the air-ratio. Experimental results show that the proposed OLSSVM can effectively predict the air-ratio to the target values under varies operating conditions and is superior to the air-ratio models available in the recent literatures. Therefore, the proposed OLSSVM is a promising scheme for automotive engine modelling.