Wide-band modeling and parameter identification in magnetostrictive actuators

Smart materials, such as magnetostictives, piezoelectrics, and shape memory alloys, all display certain coupling phenomena between applied electromagnetic fields and their mechanical properties. Smart actuators made of these materials can be used to build smart structures with the ability to sense and respond to environmental changes to achieve desired goals. However, the complicated electro-magnetic behaviors of such actuators limit effective use of them. We introduce a model which describes most of the characteristics of magnetostrictive actuators in a large frequency range, from 0-2kHz. Rather than treating each behavior separately, we treat them with arrays of similar structural parameters to minimize difficulties arising from usage of such a model for control purposes.

We develop a low frequency model with DC resistance and two current sinks, which is sufficiently accurate for frequencies less than 500Hz. Even at 500Hz, it is challenging to identify the parameters due to the nonlinearity of the hysteresis losses . To reduced this difficulty we derive a method to fix the hysteresis losses and develop a parameter identification process using sinusoidal input voltage. We then introduce an array of current sinks to improve the model for frequencies up to 800Hz. To accommodate the additional series losses that are significant after 1kHz, we introduce a transfer function which appears before the low frequency model. We experimentally verify that such models are indeed sufficiently accurate.