Application of condition monitoring to an electromechanical actuator: a parameter estimation based approach

Abstract

As part of a recent R&D programme, a model-based monitoring technique for an electromechanical actuator has been the developed and validated on a laboratory scale plant. Estimates of the key physical parameters of the electromechanical positioning system are calculated from a transfer function model of the system, which is obtained using system identification techniques. The estimates are then assessed against baseline parameters and trended historically for monitoring purposes. In addition, a fuzzy-logic based significant change detector and fault classifier is applied to the estimates. The experimental results highlight the considerable potential of the applied techniques for achieving improved condition monitoring of actuation systems.