.png){kind=icon}
.png){kind=icon}
Return
Machine learning based design optimization of a composite-inspired multiphase lattice structure for high energy absorption ability
.png){kind=icon}
PRE
AI
M
N
H
K
DOI：10.1080/15376494.2026.2696451.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this study, a novel approach is used to predict the best positions of the reinforcement phase lattice in composite-inspired multiphase lattice structures for maximum energy absorption using different machine learning (ML) models. The previous studies were based on a few random arrangements of reinforcement phase ignoring the other potential arrangements that can lead to higher energy absorption values. Multiphase lattices, also called multilattices, analyzed in this study combined BCC unit cell, which shows flat plateau stress behavior, and high-strength octet unit cell as matrix and reinforcement phases, respectively. Three different ML models, i.e. support vector regression (SVR), XGBoost, and artificial neural network (ANN), used to predict the best reinforcement positions, the ANN model exhibited highest accuracy (92.0%). The variation of reinforcement phase positions leads to significant change in energy absorption of the multilattice at same volume fraction (11.11%) of the reinforcement phase and relative density. The three maximum-energy-absorbing structures exhibited an increase in energy absorption by 85.4–83.8% compared to the parent BCC lattice and by 32.8–34.0% compared to the parent octet lattice in the FE simulations. Thus, this study demonstrates that composite-inspired multiphase lattice structure can be designed to achieve higher energy absorption ability than both parent lattices.
Keywords:
Composite-inspired metamaterial
multiphase lattice
multilattice
energy absorption
machine learning
Journal
IF:
0
Papers:
4.7K
Citations:
1.4W
Organization
Cited Papers