Preprocessing algorithms for the estimation of ordinary differential equation models with polynomial nonlinearities

Abstract The data analysis task of determining a model for an ordinary differential equation (ODE) system from given noisy solution is addressed. Since modeling with ODE ubiquitous in science and technology, finding models paramount importance. Based on previously published parameter estimation method models, four related algorithms were developed. are tested over 20 different polynomial systems comprising 60 equations at various noise levels. Two frequently compute the correct model. They compared to prominent SINDy-family those SINDy-algorithms that have simple default hyperparameters. This demonstrates they comparable SINDy more resilient towards than algorithms.