Neural Networks in Structural Control

by F. Casciati, Univ of Pavia, Pavia, Italy,
E. De Petra, Univ of Pavia, Pavia, Italy,
L. Faravelli, Univ of Pavia, Pavia, Italy,

Abstract: Let the response of a simple two-story frame to a stochastic dynamic excitation be controlled by activating suitable actuators. A neural network is trained to recognize, from the measurement of the input intensity at the beginning of the excitation, the benefit that can be obtained by the control for that specific input time history. It is a preliminary step toward the design of a neural network which decides whether or not to activate the actuators in an actively controlled structure.

Subject Headings: Frames | Excitation (physics) | Neural networks | Structural control | Stochastic processes | Structural design | Hydraulic structures | Control systems

Services: Buy this book/Buy this article


Return to search