Decentralized Discrete-Time Neural Network Controller for a Class of Nonlinear Systems with Unknown Interconnections
This document has been relocated to http://scholarsmine.mst.edu/ele_comeng_facwork/956
There were 1 downloads as of 27 Jun 2016.
A novel decentralized neural network (NN) controller in discrete-time is designed for a class of uncertain nonlinear discrete-time systems with unknown interconnections. Neural networks are used to approximate both the uncertain dynamics of the nonlinear systems and the unknown interconnections. Only local signals are needed for the decentralized controller design and the stability of the overall system can be guaranteed using the Lyapunov analysis. Further, controller redesign for the original subsystems is not required when additional subsystems are appended. Simulation results demonstrate the effectiveness of the proposed controller. The NN does not require an offline learning phase and the weights can be initialized at zero or randomly. Simulation results verify the theoretical conclusions.