An improved approach to nonlinear dynamical system identification using PID neural networks

The ability of a neural network to realize some complex nonlinear system makes it attractive for system identification. The problem of deterministic nonlinear system identification is considered as an application of PID neural network. PID neural network's weights are adjusted by the back-propagation algorithm, BFGS optimization algorithm and Levenberg-Marquardt approximation. Two examples are given to demonstrate the performance and efficiency of the proposed method. The simulation results show that the suggested method gives less effor minimization and faster convergence.