Translator Disclaimer
6 June 1997 Adaptive nonlinear neural network controller for rotorcraft vibration
Author Affiliations +
This paper presents research into developing an adaptive nonlinear neural network control algorithm that can be used with smart structure actuators and sensors to control the vibrations of rotor blades. The dynamic equations of motion for a blade have the same form as a multilink manipulator (robot arm) and adaptive nonlinear control algorithms have proven successful in active control of these manipulators. The recent development of neural network control algorithms has provided the ability to adaptively learn in real time a set of parameters that will approximate external forces operating on the blades. The controller combines these two control techniques enabling the controller to adapt its parameters in response to changes in blade properties such as its mass or stiffness and to also learn the parameters necessary to account for the unknown but bounded, periodic disturbance forces such as those caused by the unsteady, periodic aerodynamic forces in the rotor system. Current efforts have been directed at testing the control algorithm on real beams with piezoceramic actuators and sensors. The initial test results have shown that vibration reduction and desired beam motion tracking can be achieved even under the influences of periodic disturbances.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michael G. Spencer, Robert M. Sanner, and Inderjit Chopra "Adaptive nonlinear neural network controller for rotorcraft vibration", Proc. SPIE 3041, Smart Structures and Materials 1997: Smart Structures and Integrated Systems, (6 June 1997);

Back to Top