Control of a camless engine electromechanical actuator: position reconstruction and dynamic performance analysis

• Published in: IEEE transactions on industrial electronics (1982) 2004-04, Vol.51 (2), p.299-311
• Main Authors: Montanari, M., Ronchi, F., Rossi, C., Tonielli, A.
• Format: Article
• Language: English
• Subjects: Actuators; Control systems; Costs; Dynamical systems; Dynamics; Electromagnets; Electromechanical sensors; Engines; Internal combustion engines; Performance analysis; Pollution; Reconstruction; Sensitivity analysis; Studies; Torque; Trajectories; Valves
• ISSN: 0278-0046; 1557-9948
• DOI: 10.1109/TIE.2004.825230

Camless internal combustion engines offer major improvements over traditional engines in terms of efficiency, maximum torque and power, and pollutant emissions. Electromechanical valve actuators are very promising in this context, but they present significant control problems. Further, to keep system cost at an acceptable level, a control system without a valve position sensor needs to be adopted. Low valve seating velocity, small transition time for valve opening and closing, and unavailability of position sensor are conflicting objectives that need to be jointly considered. In this paper, a control system architecture is presented, capable of dealing with all these issues. It is shown that a position tracking controller is needed: a key point is the design of the reference trajectory to be tracked. Actuator physical limitations strongly influence the feasible trajectory when low valve seating velocity is required, thus affecting valve transition time. Owing to the same limitations, valve electromagnets have to be energized for a significant part of the trajectory, thus allowing valve position reconstruction starting from electrical measurements only. A method for position reconstruction is presented, which makes use of auxiliary coils to reconstruct electromagnets fluxes; it is shown via sensitivity analysis that the functional characteristics of position reconstruction and its accuracy are compatible with the required applications. The trajectory design is then addressed as an optimization problem that explicitly considers the tradeoff between fast dynamic performance and system robustness. The solution of this optimization problem enlightens the limitations on achievable dynamic performance, which are presented and discussed.