Case Study: Application of NNARX Networks in the Identification of a Small-Scale Level System

This work proposes the identification study of a small scale nonlinear system through Neural Network AutoRegressive with eXternal input (NNARX). The use of the NNARX network requires a proper methodology for its configuration and a good training set. This work proposes the main parameter definition of the network through the analysis of non-intrusive performance indexes with the support of a database based on the system response, considering a Pseudo-Random Binary Sequence (PRBS) excitation. The methodology will be applied in two specific open-loop identification situations: numerical simulation of a fourth order polynomial system (Case 01), and an experimental system that controls the water tank level with nonlinear characteristics (Case 02). The results of the identified models by the developed NNARX network were able to represent the system dynamics with high fidelity, presenting an average identification error of less than 0.14% (Case 01) and 0.34% (Case 02). It was also observed the learning and generalization evidence that could represent the process intrinsic nonlinearities satisfactorily. Besides, it will be possible to find the potentiality and usefulness of the NNARX network in non-linear system identification. Besides, it will be possible to find the potentiality and usefulness of the NNARX network in nonlinear system identification.