Deep learning-enhanced design for functionally graded auxetic lattices

Auxetic materials with counterintuitive negative Poisson’s ratio have been of significant interest due to potential
applications across diverse engineering fields. Functionally grading such auxetics further enables customization
of the structural response and harnesses the potential for multi-functional applications. However, a critical
challenge in designing functionally graded lattices is to efficiently determine the spatial variation of the functional gradient and the corresponding geometric designs to achieve the desired response. In this paper, a highly efficient deep learning-based inverse design framework for functionally graded tetra-petal auxetics with spatially tailored properties is presented. This framework significantly improves the efficiency of tailoring functionally graded auxetics where many unit cells need to be tailor-designed. The graded tetra-petal auxetics obtained from the inverse design framework are additively manufactured and subjected to impact tests. The results show superior impact performance compared with uniform designs, demonstrating the effectiveness of the proposed inverse design framework, which can be inspirable to promote advanced structures/materials with enhanced impact resistance.

Fail-safe topology optimization for multiscale structures

This paper presents a novel fail-safe topology optimization method for multiscale structures. The partial
damage of both macroscopic and microscopic scales is considered for structural design. To ensure precision,
the effective elasticity tensor obtained by the homogenization method is fitted as a high-order polynomial
function. Meanwhile, the simplified models of partially damaged truss-like microstructure are
employed to reduce the computational cost and the difficulty of fitting. Moreover, Heaviside projection
is applied to speed up the convergence and yield a relatively clear configuration. Three numerical examples
are tested to demonstrate that the optimized multiscale structures successfully obtain comprehensive
performances than optimized solid structures when appropriate microstructure configurations are
chosen. Besides, multiscale structures are more self-supporting than solid structures and thus more suitable
for additive manufacturing due to the large number of gray elements diffused

Multi-resolution topology optimization using B-spline to represent the density field

This paper proposes a novel multi-resolution topology optimization method using B-spline to represent the
density field, and overcomes the defects of tedious post-processing of element-based models and low computational efficiency of topology optimization for large-scale problems. The design domain embedded in the Bspline space is discretized with a coarser analysis mesh and a finer density mesh to reduce the computational cost of finite element analysis. As design variables, the coefficients of the control points control the shape of the Bspline. The optimized B-spline can be quickly and precisely converted into a CAD model. Sensitivity filtering is additionally applied to enhance B-spline’s smoothness and suppress QR-patterns. Numerical examples, including 2D and 3D cases, are tested to demonstrate that the proposed method significantly saves computational time without sacrificing the performance of the optimized structure. Moreover, the post-processing procedures are streamlined, resulting in continuous, smooth, and editable models.

From computer-aided design (CAD) toward human-aided design (HAD): an isogeometric topology optimization approach

In this paper, the novel design mode of human-aided design (HAD) is proposed to replace conventional
computer-aided design (CAD). In HAD, computers can automatically complete the whole product design
via a new isogeometric topology optimization (ITO), while humans just assist to slightly modify the
design to meet requirements. An embedded domain ITO is presented to design complex models with
irregular design domains, and editable geometric models of optimized results can be automatically generated
based on layered ITO results. Three examples are tested to verify the proposed HAD mode, including
a 3D cantilever beam with a regular design domain, an automotive part with an irregular design
domain, and a Messerschmitt-Bölkow-Blohm (MBB) beam with a multiscale structure. The results
demonstrate that the proposed HAD mode can automatically deliver high-quality optimized models;
thus, it has great potential as a revolutionary technology to change the current design mode from CAD
to HAD.

Open-Source Codes of Topology Optimization: A Summary for Beginners to Start Their Research

Topology optimization (TO), a numerical technique to find the optimal material layout with a given design domain, has attracted interest from researchers in the field of structural optimization in recent years. For beginners, open source codes are undoubtedly the best alternative to learning TO, which can elaborate the implementation of a method in detail and easily engage more people to employ and extend the method. In this paper, we present a summary of various open-source codes and related literature on TO methods, including solid isotropic material with penalization (SIMP), evolutionary method, level set method (LSM), moving morphable components/voids (MMC/MMV) methods, multiscale topology optimization method, etc. Simultaneously, we classify the codes into five levels, from easy to difficult, depending on their difficulty, so that beginners can get started and understand the form of code implementation more quickly

Multi-objective topology optimization filled with multiple microstructures

Structures in engineering application may face loads from multiple physical fields. To simultaneously design macroscopic structures that have lower thermal compliance and higher natural frequency, a new multi-objective topology optimization filled with multiple microstructures is proposed based on the weight sum method. To shorten the gap between the optimized results and the design requirement, a self-selected weight sum method that is based on the fitting functions of the result domains and the bisection method is proposed to get the optimized macroscopic structures with specific properties directly. Several numerical examples, including single-phase material and multiple materials cases, are presented to demonstrate the feasibility and practicality of the proposed method. The results show that the employment of multiple materials optimization provides the structures with a wider result domain than the single-phase material situations. The self-selected weight sum method is of high efficiency, good connectivity and easy to implement.