Analog Neural Network for Multivariable Algebraic Loops in Constrained Control
Abstract
Abstract
An analog neural network is designed to efficiently solve the proposed dynamic multivariable algebraic loop representation of a constrained control problem. The neural network is formulated by introducing dynamics into the static algebraic loop representation of the problem which allows it to be applied to a larger class of control problems. The proposed circuit implementation is practical and can be realized directly with passive
components and operational amplifiers or with a field programmable analog array (FPAA). An example with various MATLAB simulations is included to demonstrate the effectiveness of the solution.
Description
Department of Electrical and Computer Engineering
Rights
File access restricted due to FERPA regulations