Open AccessAutomatic differential equations identification by self-configuring genetic programming algorithm
Author(s) -
Tatiana Karaseva
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/734/1/012093
Subject(s) - genetic programming , a priori and a posteriori , identification (biology) , computer science , differential equation , algorithm , stability (learning theory) , genetic algorithm , task (project management) , point (geometry) , symbolic regression , noise (video) , sample (material) , differential (mechanical device) , reduction (mathematics) , mathematical optimization , mathematics , artificial intelligence , machine learning , engineering , mathematical analysis , philosophy , chemistry , botany , geometry , systems engineering , epistemology , chromatography , image (mathematics) , biology , aerospace engineering
The paper considers a reduction of differential equations identification problem to the symbolic regression task. The current approach allows automatic determining the structure of a differential equation via the usage of the self-configuring genetic programming algorithm. The a priori information needed is only the dynamic system initial point and the sample of input and output effects. The stability of the proposed approach to the presence of noise in the sample and the small amount of data is investigated.