View article

This paper describes a so-called ‘Adaptive Modelling’procedure by which the TWR convergence is improved for nonlinear systems. The procedure adapts the linear system model during the TWR iterations by comparing the predicted and measured drive and/or response signals. The improved performance of the proposed procedure is demonstrated on a virtual test rig which consists of a multi-body simulation model of a vehicle, whereby real-life measured spindle forces are used as target definition.