Topologically optimal design and failure prediction using conditional generative adversarial networks

Abstract

Among the various structural optimization tools, topology optimization is the widely used technique in obtaining the initial design of structural components. The resulting topologically optimal initial design will be the input for subsequent structural optimizations such as shape, size and layout optimizations. However, iterative solvers used in conventional topology optimization schemes are known to be computationally expensive, thus act as a bottleneck in the manufacturing process. In this paper, a novel deep learning‐based accelerated topology optimization technique with the ability to predict ductile material failure is presented. A Conditional Generative Adversarial Network (cGAN) coupled with a Convolutional Neural Network (CNN) is used to predict the optimal topology of a given structure subject to a set of input variables. Subsequently, the same cGAN is trained to predict the Von‐Mises stress contours on the optimal structure by means of color transformed image‐to‐image translations. The ductile failure criterion is evaluated by comparing the cGAN predicted maximum Von‐Mises stress with the yield strength of the material. The proposed novel numerical method is proven to arrive at the topologically optimal design, accompanying the material failure decision within a negligible amount of time but also maintaining a higher prediction accuracy.

A finite element-convolutional neural network model (FE-CNN) for stress field analysis around arbitrary inclusions

“…The results showed that increasing the resolution led to improved prediction accuracy, as indicated by the correlation values presented in table 1. This agrees with similar models from the literature [72,73] in which higher resolution led to increased accuracy. Figure 15 shows the training and testing loss of the three models trained with different resolutions.…”

Nonlinear response modelling of material systems using constrained Gaussian processes

“…These optimal designs are derived after the systematic processing through one or more of the topology, size, shape, and layout optimisations. 60 The primary motivation for choosing this problem lies in the expected geometric nonlinear behaviour of the optimal structure due to its relative slenderness. The accurate CGPR prediction of the force-displacement behaviour of the optimal structure is crucial for design purposes as optimisation algorithms are generally expensive.…”

“…Structural optimisation of components and assemblies has gained increasing popularity due to the high stiffness‐to‐weight ratio of the optimal design. These optimal designs are derived after the systematic processing through one or more of the topology, size, shape, and layout optimisations 60 . The primary motivation for choosing this problem lies in the expected geometric nonlinear behaviour of the optimal structure due to its relative slenderness.…”

Real‐time topology optimization via learnable mappings

“…For instance, in Reference 20, the boundary conditions, loading conditions, volume fraction and domain size are directly encoded into the input tensor of a U‐Net autoencoder (AE) architecture that is exploited to predict the corresponding optimal topology. An alternative solution was proposed in Reference 21, through the combined use of CNNs and conditional generative adversarial networks, which can generate the optimal design and trace the corresponding von Mises (VM) stress contour, eventually allowing the identification of the most stressed regions.…”