pyVPLM (Variable Power-Law regression Models)
pyVPLM is a package that is developed to help scientist, engineer, etc., to construct power-law and/or polynomial regression models on different type of data such as finite-element simulation results, manufacturer data-sheets ... It integrates various functionalities such as :
- Model parameters reduction based on Buckingham Theorem dimensional analysis and Pint package with derived functions.
- Sensitivity and dependency analysis on dimensionless parameter and limited experiments to simplify further model expressions.
- Construction of optimized experimental design on feasible-physical variables leading to full-factorial design within dimensionless space. Those DOE are the inputs of parametrized finite-element models.
- Regression models construction with increasing complexity (terms sorted based on their impact) and validation based on relative error repartition analysis.
Download : https://github.com/SizingLab/pyvplm
Documentation : https://pyvplm.readthedocs.io
FAST-OAD (Future Aircraft Sizing Tool - Overall Aircraft Design)
FAST-OAD is a framework for performing rapid Overall Aircraft Design. It proposes multi-disciplinary analysis and optimisation by relying on the OpenMDAO framework. FAST-OAD allows easy switching between models for a same discipline, and also adding/removing disciplines to match the need of your study. Currently, FAST-OAD is bundled with models for commercial transport aircraft of years 1990-2000. Other models will come, and you may create your own models and use them instead of bundled ones.
Download : https://github.com/fast-aircraft-design/FAST-OAD
Documentation : https://fast-oad.readthedocs.io/en/v1.0.5/
SMT (Surrogate Modeling Toolbox)
SMT is a Python package that contains a collection of surrogate modeling methods, sampling techniques, and benchmarking functions. This package provides a library of surrogate models that is simple to use and facilitates the implementation of additional methods. SMT is different from existing surrogate modeling libraries because of its emphasis on derivatives, including training derivatives used for gradient-enhanced modeling, prediction derivatives, and derivatives with respect to the training data. It also includes new surrogate models that are not available elsewhere: kriging by partial-least squares reduction and energy-minimizing spline interpolation. SMT is documented using custom tools for embedding automatically-tested code and dynamically-generated plots to produce high-quality user guides with minimal effort from contributors.