Chapter 4 - Machine learning-enabled parametrically upscaled constitutive models for bridging length scales in Ti and Ni alloys

Created by W.Langdon from gp-bibliography.bib Revision:1.8414

@InCollection{GHOSH:2024:ILHA,

• author = "Somnath Ghosh",
• title = "Chapter 4 - Machine learning-enabled parametrically upscaled constitutive models for bridging length scales in Ti and Ni alloys",
• booktitle = "Innovative Lightweight and High-Strength Alloys",
• publisher = "Elsevier",
• year = "2024",
• editor = "Mohammed A. Zikry",
• pages = "97--139",
• keywords = "genetic algorithms, genetic programming, Parametrically upscaled constitutive model, Parametrically upscaled crystal plasticity model, Dual-phase titanium alloy, Ni-based superalloys, Machine learning, ANN",
• isbn13 = "978-0-323-99539-9",
• URL = "https://www.sciencedirect.com/science/article/pii/B9780323995399000047",
• DOI = "doi:10.1016/B978-0-323-99539-9.00004-7",
• abstract = "This chapter provides an overview of the development of the parametrically upscaled constitutive model (PUCM) for Ti alloys like Ti-6AL-4V, and the parametrically upscaled crystal plasticity model (PUCPM) for single crystals in Ni-based superalloys. These thermodynamically consistent constitutive models bridge multiple spatial scales through the explicit representation of representative aggregated microstructural parameters. They enable computationally efficient simulations with significant speedup over detailed lower scale models. A host of computational tools and machine learning algorithms are developed to create an automated pipeline for parametric upscaling. The novel algorithms used include genetic algorithm with support vector regression, Sobol analysis-based global sensitivity analysis, artificial neural network for emulating the crystal plasticity finite element models, nonlinear optimisation scheme using k-means acceleration, and genetic programming symbolic regression for functional representation. The computational tool chain outputs the highly efficient PUCMs/PUCPMs, which are invaluable tools for multiscale analysis of deformation and failure with implications in location-specific design",

}

Genetic Programming entries for Somnath Ghosh

Citations