Neural-Net-Control

This respository contains the code for the identification and control of nonlinear dynamical systems using Neural Networks. The examples for this experiment is taken from 'Identification and Control of Dynamical Systems using Neural Networks' by K. S. Narendra and Kannan Parthasarathy.

There are 3 different implementations for the same:

1. BPA - This is the implementation using simple back propagation with 2 hidden layers.
2. BPA - Single Layer - This uses only a single hidden layer.
3. OSLA - This is the implentation of the Online Sequential Learning Algorithm.

All the above results can be compared using the MSE (mean square error) and the VAF (variance accounted for) parameters.