Cycle Adaptive Feedforward Approach Controllers for an Electromagnetic Valve Actuator

摘要: An electromagnetic valvetrain for an internal combustion engine can improve the thermal efficiency but requires soft landing control to avoid excessive wear and acoustic noise. To simplify problem, valve algorithm is partitioned into approach controllers. The responsible last part of 8 mm travel while rest. This paper focuses on control. goal achieve end state that sets constant initial conditions In addition identified system model, information based previous events utilized take advantage repetitive nature opening/closing. Nonlinear iterative learning, terminal Nelder Mead direct search algorithms are three cyclic adaptive feedforward controllers tested in simulation automotive valves. These results compared emphasis this which works well both experiment disturbances slow time.