Approach control of an electromechanical valve actuator using closed-loop iterative learning control

Electromechanical valves (EMV) are a promising alternative to camshafts in internal combustion (IC) engines. This paper is mainly concerned with the control of EMVs to achieve a soft landing. Due to the nature of the process, control of the valve is normally accomplished in two phase. This paper addresses the control problem over the first phase, namely the approach control , where the control authority over the system is very low. Repetitiveness of the process and low control authority suggest iterative learning control (ILC) as an appealing solution. An ILC law that exploits the benefits of both feedforward and feedback approaches is proposed for this problem. Also, an iterative method for characterization and generation of the desired trajectory is provided. The effectiveness of the proposed approach controller is demonstrated through simulation studies. © 2011 IEEE