Reconstruction of observed mechanical motions with artificial intelligence tools

作     者：Jakovac, Antal Kurbucz, Marcell T. Posfay, Peter 

作者机构：Wigner Res Ctr Phys Dept Computat Sci 29-33 Konkoly Thege Miklos St H-1121 Budapest Hungary Corvinus Univ Budapest Dept Stat 8 Fovam Sq H-1093 Budapest Hungary 

出 版 物：《NEW JOURNAL OF PHYSICS》 (New J. Phys.)

年 卷 期：2022年第24卷第7期

页      面：073021页

核心收录：

学科分类：07[理学] 0702[理学-物理学] 

基　　金：Ministry of Innovation and Technology NRDI Office Hungarian Research Fund NKFIH (OTKA) [K123815] 

主　　题：data driven modeling mechanical motions artificial intelligence numerical determination of physical laws renormalization 

摘      要：The goal of this paper is to determine the laws of observed trajectories assuming that there is a mechanical system in the background and using these laws to continue the observed motion in a plausible way. The laws are represented by neural networks with a limited number of parameters. The training of the networks follows the extreme learning machine idea. We determine laws for different levels of embedding, thus we can represent not only the equation of motion but also the symmetries of different kinds. In the recursive numerical evolution of the system, we require the fulfillment of all the observed laws, within the determined numerical precision. In this way, we can successfully reconstruct both integrable and chaotic motions, as we demonstrate in the example of the gravity pendulum and the double pendulum.