Parameter Identification for Electrohydraulic Valvetrain Systems

We consider an electrohydraulic valve system (EHVS) model with uncertain parameters that may possibly vary with time. This is a nonlinear third order system consisting of two clearly separated subsystems, one for the piston position and the other for the chamber pressure. The nonlinearities involved are flow-pressure characteristics of the solenoid valves, the pressure dynamics of the chamber due to varying volume, and a variable damping nonlinearity. We develop a parametric model that is linear in the unknown parameters of the system using filtering. We deal with a nonlinear parameterization in the variable damping term using the Taylor approximation. We design a parameter identifier, which employs a continuous-time unnormalized least-squares update law with a forgetting factor. This update law exponentially converges to the true parameters under a persistence of excitation condition, which is satisfied due to the periodic regime of operation of EHVS. We present simulation results that show good following of unknown parameters even with the presence of sensor noise.