|
View: |
Part 1: Document Description
|
|
Citation |
|
|---|---|
|
Title: |
ApHIN - Autoencoder-based port-Hamiltonian Identification Networks (Software Package) |
|
Identification Number: |
doi:10.18419/darus-4446 |
|
Distributor: |
DaRUS |
|
Date of Distribution: |
2024-08-27 |
|
Version: |
1 |
|
Bibliographic Citation: |
Kneifl, Jonas; Rettberg, Johannes; Herb, Julius, 2024, "ApHIN - Autoencoder-based port-Hamiltonian Identification Networks (Software Package)", https://doi.org/10.18419/DARUS-4446, DaRUS, V1 |
|
Citation |
|
|
Title: |
ApHIN - Autoencoder-based port-Hamiltonian Identification Networks (Software Package) |
|
Identification Number: |
doi:10.18419/darus-4446 |
|
Identification Number: |
swh:1:snp:6e04af96fced0103820f0ce62035fc3a5d00523e;origin=https://github.com/Institute-Eng-and-Comp-Mechanics-UStgt/ApHIN |
|
Authoring Entity: |
Kneifl, Jonas (University of Stuttgart) |
|
Rettberg, Johannes (University of Stuttgart) |
|
|
Herb, Julius (University of Stuttgart) |
|
|
Grant Number: |
EXC 2075 - 390740016 |
|
Grant Number: |
Artificial Intelligence Software Academy (AISA) |
|
Grant Number: |
InnovationsCampus Future Mobility |
|
Distributor: |
DaRUS |
|
Access Authority: |
Kneifl, Jonas |
|
Access Authority: |
Fehr, Jörg |
|
Depositor: |
Kneifl, Jonas |
|
Date of Deposit: |
2024-08-15 |
|
Holdings Information: |
https://doi.org/10.18419/DARUS-4446 |
|
Study Scope |
|
|
Keywords: |
Computer and Information Science, Engineering, Mathematical Sciences, Physics, System Identification, Port-Hamiltonian Systems, Reduced-Order Modeling, Structure-Preserving, Autoencoder |
|
Topic Classification: |
Computer Science |
|
Abstract: |
<h3>Software package for data-driven identification of latent port-Hamiltonian systems.</h3> <h3>Abstract</h3> Conventional physics-based modeling techniques involve high effort, e.g.~time and expert knowledge, while data-driven methods often lack interpretability, structure, and sometimes reliability. To mitigate this, we present a data-driven system identification framework that derives models in the port-Hamiltonian (pH) formulation. This formulation is suitable for multi-physical systems while guaranteeing the useful system theoretical properties of passivity and stability.<br> Our framework combines linear and nonlinear reduction with structured, physics-motivated system identification. In this process, high-dimensional state data obtained from possibly nonlinear systems serves as the input for an autoencoder, which then performs two tasks: (i) nonlinearly transforming and (ii) reducing this data onto a low-dimensional manifold. In the resulting latent space, a pH system is identified by considering the unknown matrix entries as weights of a neural network. The matrices strongly satisfy the pH matrix properties through Cholesky factorizations. In a joint optimization process over the loss term, the pH matrices are adjusted to match the dynamics observed by the data, while defining a linear pH system in the latent space per construction. The learned, low-dimensional pH system can describe even nonlinear systems and is rapidly computable due to its small size.<br> The method is exemplified by a parametric mass-spring-damper and a nonlinear pendulum example as well as the high-dimensional model of a disc brake with linear thermoelastic behavior <h3>Features</h3> This package implements neural networks that identify linear port-Hamiltonian systems from (potentially high-dimensional) data [1]. <ol> <li>Autoencoders (AEs) for dimensionality reduction <li>pH layer to identify system matrices that fullfill the definition of a linear pH system <li>pHIN: identify a (parametric) low-dimensional port-Hamiltonian system directly <li>ApHIN: identify a (parametric) low-dimensional latent port-Hamiltonian system based on coordinate representations found using an autoencoder <li>Examples for the identification of linear pH systems from data <ul> <li>One-dimensional mass-spring-damper chain <li>Pendulum <li>discbrake model </ul > <br></li> </ol> See <a href="https://institute-eng-and-comp-mechanics-ustgt.github.io/ApHIN/">documentation</a> for more details. |
|
Methodology and Processing |
|
|
Sources Statement |
|
|
Data Access |
|
|
Other Study Description Materials |
|
|
Related Publications |
|
|
Citation |
|
|
Title: |
Johannes Rettberg, Jonas Kneifl, Julius Herb, Patrick Buchfink, Jörg Fehr, and Bernard Haasdonk. Data-driven identification of latent port-Hamiltonian systems. Arxiv, 2024. |
|
Identification Number: |
2408.08185 |
|
Bibliographic Citation: |
Johannes Rettberg, Jonas Kneifl, Julius Herb, Patrick Buchfink, Jörg Fehr, and Bernard Haasdonk. Data-driven identification of latent port-Hamiltonian systems. Arxiv, 2024. |
|
Label: |
ApHIN.zip |
|
Notes: |
application/zip |