Iterative learning control of electromechanical camless valve actuator

In this paper we consider the control of the valve trajectory in a gasoline engine equipped with an electromechanical camless valvetrain (EMCV) actuator. Precise control of the voltage applied to the coils is required to avoid high contact velocities during valve closing and opening, i.e., to ensure soft landing. In our previous work we designed a tracking controller that consists of a constant preset voltage augmented by a voltage command based on a linear feedback and modified by an Iterative Learning Controller (ILC). The new results reported here involve the preset voltage in the learning process. This yields good results in the presence of unknown gas flow forces through the valves. As a consequence, a non-square learning algorithm is used to calculate more input values than available error values. We demonstrate the closed loop performance through simulations with an unknown acting gas flow force that emulates a realistic varying engine load