Cyclic RGF regulation using adaptive IMC approach and statistical feedback criterion

Abstract

Abstract Residual gas fraction (RGF) significantly affects fuel efficiency and emission of combustion engines, and usually be regulated by managing variable valve timing (VVT) system. Cyclic RGF regulation is challenging due to nonlinear parameter-varying dynamics as well as stochastic property of controlled objects. This paper addresses this issue by designing an adaptive internal model controller (AIMC) combined with statistical feedback criterion. Firstly, closed-loop RGF regulation problem based on VVT control is formulated. Secondly, AIMC is adopted to online identify plant dynamics and its inverse model. Note that the statistical criterion is creatively introduced in the feedback channel of AIMC to handle the stochastic property of RGF. Finally, experiments performed on an engine test bench show that the proposed controller not only possesses advantages of AIMC but also alleviates stochastic noise in feedback channel.