Skip to main content

AI-powered framework to solve both forward and inverse problems in structural dynamics

Project description

AIStructDynSolve

Author:

AIStructDynSolve is an artificial intelligence (AI) powered framework designed to solve both forward and inverse problems in structural dynamics. It leverages advanced artificial intelligence methods - particularly physics-informed neural networks (PINNs), Physics-Informed Kolmogorov-Arnold Network(PIKANs) and their extensions - to model, predict, and analyze dynamic structural responses under various loading scenarios, such as seismic excitations.

The framework solves the following ODE of MDOF:

M*U_dotdot+C*U_dot+K*U=Pt

  • Initial Conditions:
  • U(t=0)=InitialU
  • U_dot(t=0)=InitialU_dot

The framework aims to:

  • Accurately simulate time-dependent structural behavior (forward problems).
  • Identify structural parameters or input forces from measured responses (inverse problems).
  • Incorporate domain knowledge and physical laws for improved generalization and interpretability.
  • Address challenges in multi-frequency, multi-scale dynamics, especially in earthquake engineering applications.

Date:

  • 2023/12/26

Update Log

  • Version 1.10 - May 11, 2025 Improved accuracy and stability of PIKANs

  • Version 1.9 - Apr 30, 2025 Added support for EQ2D and EQ3D of structural dynamic simulations, enabling users to analyze 2D and 3D earthquake loading scenarios

  • Version 1.8 - Apr 12, 2025 Improved accuracy of high-frequency response prediction

  • Version 1.7 - Apr 11, 2025 Minor bug fixes

  • Version 1.6 - Apr 7, 2025 Minor bug fixes

  • Version 1.5 - Jan 13, 2025 Minor bug fixes

  • Version 1.4 - Dec 31, 2024 Minor bug fixes

  • Version 1.3 - Dec 19, 2024 Minor bug fixes

  • Version 1.2 - Dec 18, 2024 Minor bug fixes

  • Version 1.1 - Dec 7, 2024 Minor bug fixes

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aistructdynsolve-1.10.tar.gz (33.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

AIStructDynSolve-1.10-py3-none-any.whl (83.4 kB view details)

Uploaded Python 3

File details

Details for the file aistructdynsolve-1.10.tar.gz.

File metadata

  • Download URL: aistructdynsolve-1.10.tar.gz
  • Upload date:
  • Size: 33.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.12.7

File hashes

Hashes for aistructdynsolve-1.10.tar.gz
Algorithm Hash digest
SHA256 2006c29552c83693d7e31e8bbdf6bd2bf8260043e44bf38f46ec15bd81b987b8
MD5 a71caab9f072e88772f74b6a06d72098
BLAKE2b-256 d281e594a20d3e08625a6c47bb1039af7dff7740dc837f595a41f9c3eb7848cc

See more details on using hashes here.

File details

Details for the file AIStructDynSolve-1.10-py3-none-any.whl.

File metadata

File hashes

Hashes for AIStructDynSolve-1.10-py3-none-any.whl
Algorithm Hash digest
SHA256 30d4f5ce1398f38081613e8dd02a86b6ca7a7e1fbcda83e5dcbe410b9988ff18
MD5 ab1470b255f57a87ac93e0f7ff64236e
BLAKE2b-256 b4c12eb0440a527cec4e700647a4be626bbd3d6f1a7304917ce883d64e46c824

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page