Welcome to PARANSYS’s documentation!

PARANSYS: Python pArametric Reliability Analysis on ANSYS, but it’s not limited just for that!

https://zenodo.org/badge/190074859.svg

PARANSYS is a module that can connect the ANSYS software to Python scripts using APDL scripts. It has one connection class, that can be used with many types of APDL analysis (more than reliability analysis), and two other classes for reliability analysis, that can be used with implicit and explicit Python functions, not just with ANSYS.

Introduction

This module calls ANSYS by Probabilistic Design System (PDS) as an FEM tool for running a couple of parametric analysis and get some variables back to Python.

The class paransys.ANSYS is the heart of the process, this class receive the ANSYS APDL model, the parameters names and values, runs everything and return the control parameters values for each simulation back to Python. This class makes easy to run a lot of analysis with the same model just changing some parameters values, in this scope we have Reliability Analysis, Optimizations and Mesh Convergence Studies, for example.

For Reliability Analysis there are two classes implemented:

  • paransys.MonteCarlo

    This class performs Monte Carlo reliability analyses, with and without using importance sampling. When using importance sampling it’s possible to fix the sampling point or search it based on the center of failures weights. It’s currently limited to one limit state.

  • paransys.FORM

    This class applies the First Order Reliability Method (FORM) inside Python using ParAnsys as a connection with ANSYS for evaluate FEM models. It is also possible to run simulations without using ANSYS, just defining the limit state equation and all variables.

How to install

PARANSYS is available at github.com/dutitello/parAnsys. You can install it by cloning (downloading) this repository and then running command python setup.py install.

How to cite

This module is published at Zenodo.

You can cite it as follows, or you can get another format in it’s Zenodo page.

Titello, E.P. (2020). PARANSYS: Python pArametric Reliability Analysis on ANSYS. Version 0.1. Zenodo. https://doi.org/10.5281/zenodo.4161580

Final notes

This is part of my master degree thesis, available here.

This module makes no claim to own any rights to ANSYS, it’s just an interface to call the program owned by ANSYS company.

Indices and tables