On system-oriented modeling and identification of magnetic shape memory (MSM) actuators

The magnetic shape memory (MSM) alloys are interesting candidates among active materials used in the actuators, particularly due to a macroscopic relationship between the applied magnetic field and the resulting MSM strain. A system-oriented description of MSM actuators is quite challenging caused by the inherently nonlinear hysteretic behavior of the MSM transducer. This paper describes a modeling and identification approach which combines the second-order linear actuator dynamics with the novel two-inputs nonlinear MSM model. The proposed identification strategy allows to decompose the linear and nonlinear effects observable in the actuator response under certain excitation conditions. The experimental evaluation performed on a prototypic MSM actuator reveals the model suitability, particulary by capturing the state-dependent memory phenomena of MSM hysteresis.