Estimating Cutting Forces in Micromilling by Input Estimation from Closed-loop Data

Estimation of cutting forces in micromilling from the signals of a spindle with Active Magnetic Bearings (AMB) is treated as an input estimation problem. For the closed-loop AMB system, a minimum mean square error input estimator with an adjustable delay is derived. This filter is based on the Wiener filter, where the unknown input is treated as white noise filtered by known dynamics and the controller is assumed to be known. It is shown that controller knowledge can be replaced by a perfect measurement of the control signal. Simulation results demonstrate the applicability of the presented approach.