Corpus ID: 221655441

Accelerating gradient-based topology optimization design with dual-model neural networks

  title={Accelerating gradient-based topology optimization design with dual-model neural networks},
  author={Chaojun Qian and Wenjing Ye},
Topology optimization (TO) is a common technique used in free-form designs. However, conventional TO-based design approaches suffer from high computational cost due to the need for repetitive forward calculations and/or sensitivity analysis, which are typically done using high-dimensional simulations such as Finite Element Analysis (FEA). In this work, neural networks are used as efficient surrogate models for forward and sensitivity calculations in order to greatly accelerate the design… Expand
Scalable Deep-Learning-Accelerated Topology Optimization for Additively Manufactured Materials
A general scalable deep-learning (DL) based TO framework, referred to as SDL-TO, which utilizes parallel schemes in high performance computing (HPC) to accelerate the TO process for designing additively manufactured (AM) materials and significantly reduces the computational cost. Expand


Kriging-assisted topology optimization of crash structures
Compared to the state-of-the-art Covariance Matrix Adaptation Evolution Strategy (CMA-ES), the KG-LSM optimization algorithm demonstrates to be efficient in terms of convergence speed and performance of the optimized designs. Expand
Multiscale topology optimization using neural network surrogate models
Because the derivative of the surrogate model is important for sensitivity analysis of the macroscale topology optimization, a neural network training procedure based on the Sobolev norm is described, and an alternative method is developed to enable creation of void regions. Expand
Parallel framework for topology optimization using the method of moving asymptotes
The complexity of problems attacked in topology optimization has increased dramatically during the past decade. Examples include fully coupled multiphysics problems in thermo-elasticity,Expand
A Topology Optimization Design for the Continuum Structure Based on the Meshless Numerical Technique
In this paper,the meshless radial point interpolation method(RPIM) is applied to carry out the topology optimization of the continuum structure.Considering the relative density of nodes as designExpand
Efficient structure topology optimization by using the multiscale finite element method
The computational accuracy and efficiency of the MsFEM method is investigated in detail, as well as the speedup ratio and parallel efficiency when using multiple processors to construct the multiscale shape functions simultaneously. Expand
Generating optimal topologies in structural design using a homogenization method
Optimal shape design of structural elements based on boundary variations results in final designs that are topologically equivalent to the initial choice of design, and general, stable computationalExpand
A topology optimization method for design of negative permeability metamaterials
A methodology based on topology optimization for the design of metamaterials with negative permeability is presented, based on a thin layer of copper printed on a dielectric, rectangular plate of fixed dimensions. Expand
Design of materials using topology optimization and energy-based homogenization approach in Matlab
This paper presents a Matlab code for the optimal topology design of materials with extreme properties. For code compactness, an energy-based homogenization approach is adopted rather than theExpand
Robust topology optimization of vibrating structures considering random diffuse regions via a phase-field method
A robust topology optimization method for structural dynamic problems by considering random diffuse-region widths between different material phases using a phase-field model, which generates meaningful optimal topologies for structuralynamic robust optimization problems with the framework of the phase- field method. Expand
A level set method for structural topology optimization
This paper presents a new approach to structural topology optimization. We represent the structural boundary by a level set model that is embedded in a scalar function of a higher dimension. SuchExpand