Incorporating Numerical Uncertainties for Validation of Nonlinear Models

TL;DR

This paper introduces a validation approach for nonlinear models that accounts for numerical errors, demonstrating significant differences in validation indexes after few iterations for specific systems.

Contribution

It presents a novel validation method incorporating numerical uncertainties, highlighting differences in model indexes not previously reported.

Findings

Significant differences (~34%) in validation indexes after few iterations.

Validation indexes like RMSE and MAPE are affected by numerical errors.

The method improves accuracy in nonlinear model validation.

Abstract

This paper proposes a model validation method that incorporates error due to numerical procedures. Two identified models for Sine Map and Duffing-Ueda Circuit systems have been investigated. The indexes RMSE and MAPE have been applied. We have shown that after some few iterates, it is possible to notice some significative difference between index provided in the literature. This difference has been computed in around 34%.