Control Design for Nonlinear Smart Actuators

North Carolina State University, Raleigh NC

Investigators

Abstract

This investigation focuses on the analysis and design of nonlinear actuators which utilize smart material transducers comprised of piezoelectric, electrostrictive or magnetostrictive materials. These transducers have proven highly effective in a number of industrial, automotive, defense, aerospace and aeronautical applications due to their dual capability for sensing and actuating, the magnitude of generated forces, and their broadband nature. At high drive levels, however, all of these materials exhibit constitutive nonlinearities and varying degrees of hysteresis. The goal in this investigation is to develop mathematical models and control algorithms which, in conjunction with drive electronics, compensate for the nonlinearities inherent to these materials. The strategy in the modeling component focuses on utilizing energy mechanisms to develop unified models that are appropriate for each of the smart material compounds. Inverse compensators based on these models will then be combined with feedback, feedforward, adaptive and robust control laws to provide the control algorithms necessary to achieve linear transducer dynamics. This investigation will advance the state of smart material control systems and extend the capabilities of currently available actuators for high performance applications by facilitating the construction of actuators which have the full stroke and force capabilities of the nonlinear constituent smart materials but exhibit linear behavior throughout their drive range.