No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Topology optimization of continuum structures under hybrid additive-subtractive manufacturing constraints
Abstract
Additive manufacturing (AM) makes it possible to fabricate complicated parts that are otherwise difficult to manufacture by subtractive machining. However, such parts often require temporary support material to prevent the component from collapsing or warping during fabrication. The support material results in increased material consumption, manufacturing time and clean-up costs. The surface precision and dimensional accuracy of the workpieces from AM are far from the engineering requirement due to layer upon layer manufacturing. Subtractive machining (SM), by contrast, can fabricate parts to satisfy the requirements of surface precision and dimensional accuracy. Nevertheless, the components need to be relatively uncomplicated for subtractive manufacturing. Thus, hybrid additive-subtractive manufacturing (HASM) is gaining increasing attention in order to take advantages of both processes. There is little research on the topology design methodology for this hybrid manufacturing technology. To address this issue, a method based on geometry approach for topology optimization of continuum structure is proposed in this paper. Both additive manufacturing and subtractive machining constraints are simultaneously considered in each topology optimization iteration. The topology optimization is performed by the bi-directional evolutionary structural optimization (BESO) method. The effectiveness of the method is demonstrated by several 3D compliance minimization problems. Fig. 1. (a) Schematic description of the HASM process. (b) Part fabricated by AM. (c) Part fabricated by HASM (Du et al. 2018). References Du W, Bai Q, Zhang B (2018) Machining characteristics of 18Ni-300 steel in additive/subtractive hybrid manufacturing. Int J Adv Manuf Technol 95(5-8):2509-2519
... Thus, research is recently focusing on integrating both constraints into TO's formulation. Zhang and Zhou (2018), Han et al. (2019), Bi, Tran and Xie (2020) adapted TO approaches to account for overhang limitations and deliver self-supporting designs. Xu et al. (2020) integrated AM support structure and thin feature constraints into Bidirectional Evolutionary Structural Optimization to obtain AM friendly designs. ...
Despite the freedom Additive Manufacturing (AM) offers when manufacturing organic shapes, it still requires some geometrical criteria to avoid a part's collapse during printing. The most synergetic design approach to AM is Topology Optimization (TO), which finds an optimal free-form given mechanical constraints. However, it is hard for TO to integrate these layout geometry-related constraints and it seldom proposes printable shapes. Therefore, this work leverages the Deep Learning (DL) capability to handle spatial correlations within the mechanical design process by integrating the layout and mechanical constraints at the conceptual level. It proposes a DL-layout-driven solution (DL-TO) trained via a triple-discriminator Generative Adversarial Network (GAN) framework. The DL-TO's performance is demonstrated by generating mechanically valid 2D designs conforming with layout constraints in a fraction of a second. DL-TO's creativity is illustrated by its capability to generate designs with unseen input constraints (passive/active elements) and to propose several shapes for the same input mechanical constraints, a task that is hard for a traditional TO to achieve.
... They generalized and improved their previous work to consider directional-dependent overhang constraint and minimum length control and generated high-resolution 3D structures [34]. Also, the authors of [6,12,18,33] adapted TO approaches to include overhang limitations and deliver self-supporting and print-ready designs. Yoely et al. [31] proposed an optimization approach constraining the areas of holes and curvatures of boundaries using the B-spline representation to obtain a manufacturable design. ...
We build an original synthetic dataset of 2D mechanical designs alongside their mechanical and geometric constraints, GMCAD. Such a dataset allows training Deep Learning (DL) models for Design for Additive Manufacturing (DfAM) to incorporate and control Computer-Aided-Design (CAD) features with mechanical performance. Geometric AM constraints are often complex to describe, depending on applications, processes, materials. They often lack explicit mathematical descriptions, belong exclusively to the CAD world, and hardly can be integrated into mechanical design, hampering AM design freedom. DL models have recently emerged as a potential to reconcile both CAD and Computer Aided-Engineering (CAE) worlds. They derive data-driven geometric rules over mechanical designs, allowing fine-grained control over the geometry during the design phase, contrary to the conventional CAD-to-CAE sequential approach. DL models, however, need high-quality labeled data, and merging CAD features to CAE aspects is challenging as they rely on different formats, rules, and tools. GMCAD dataset solves this issue following these building steps. (i) Building a DL-mechanical conditions predictor from a dataset generated by a density-gradient-based Topology Optimization method (TO); an AM-synergetic design generation tool. (ii) Creating a CAD dataset inspired by the TO-based designs. (iii) Predicting the mechanical conditions of the CADs using the DL predictor of mechanical conditions. Last, we evaluate the mechanical performance of GMCAD’s designs and derive statistics over CAD and CAE features. Designs of GMCAD show the significant influence of minor geometric changes, explaining the intricate design task of conforming both with functionality and geometric constraints. Consequently, having GMCAD is advantageous to train DL models to generate designs accounting for all these constraints simultaneously, without the need for time-consuming trial and error techniques. Such models could enhance DfAM and go beyond AM; they can also enhance other challenging fields as CAD automatic reconstruction, reverse engineering, isogeometric design and paves the way to multi-objective controllable design generation.
... Compared to HIP and incorrectly executed heat-treatment, no disadvantages exist besides the removal of some additional material. Furthermore, hybrid processes enable extensive possibilities for topology optimization 35 . ...
Additive manufacturing (AM) enables the cost-effective production of complex components, many of which are traditionally manufactured using costly production steps among other processes. One widely applied AM process is Laser-based Powder Bed Fusion of Metals (PBF-LB/M); however, internal pores and rough surfaces are typically inevitable with PBF-LB/M, reducing fatigue and corrosion resistance compared to traditional processes involving turning and milling. Additionally, large defects often occur near to or just at the surfaces. Thus, this study investigates the effect of hybrid additive and subtractive manufacturing on the fatigue strength of AISI 316L. For this purpose, different post treatment routes are compared with wrought material. Additionally, computer tomography is used to determine the necessary machining depth of the surface layer. In this study, heat-treatment and machining are both found to significantly increase fatigue strength. Finally, cyclic mean stresses affect wrought and AM specimens differently.
Additive manufacturing (AM) processes have proven to be a perfect match for topology optimization (TO), as they are able to realize sophisticated geometries in a unique layer-by-layer manner. From a manufacturing viewpoint, however, there is a significant likelihood of process-related defects within complex geometrical features designed by TO. This is because TO seldomly accounts for process constraints and conditions and is typically perceived as a purely geometrical design tool. On the other hand, advanced AM process simulations have shown their potential as reliable tools capable of predicting various process-related conditions and defects. Thus far, geometry design by topology optimization and multiphysics manufacturing simulations have been viewed as two mostly separate paradigms, whereas one should really conceive them as one holistic computational design tool. More specifically, AM process models provide input to physics-based TO, where consequently, not only the designed component will function optimally, but also will have near-to-minimum manufacturing defects. In this regard, we aim at giving a thorough overview of holistic computational design tool concepts applied within AM. First, literature on TO for performance optimization is reviewed and then the most recent developments within physics-based TO techniques related to AM are covered. Process simulations play a pivotal role in the latter type of TO and serve as additional constraints on top of the primary end-user optimization objectives. As a natural consequence of this, a comprehensive and detailed review of non-metallic and metallic additive manufacturing simulations is performed, where the latter is divided into micro-scale and deposition-scale simulations. Material multi-scaling techniques, which are central to the process-structure-property relationships, are reviewed next, followed by a subsection on process multi-scaling techniques, which are reduced-order versions of advanced process models and are incorporable into physics-based TO due to their lower computational requirements. Finally the paper is concluded and suggestions for further research paths discussed.
Background: Additive manufacturing (AM) oriented topology optimisation has become one of the most important branches in Design for Additive Manufacturing (DfAM). Traditional topology optimisation algorithms are finite element (FE) based which result in mesh-dependent zigzag or blurry boundaries, requiring extra effort to obtain accurate boundary information for manufacturing. Currently, introducing additional support structures and designing self-supporting structures are two effective ways to avoid collapse during fabrication. Self-supporting design is becoming preferable as it can reduce the use of materials and avoid the extra efforts of removing support structures after fabrication. Therefore, smooth design of self-supporting topologies is a promising research field in terms of generating print-ready geometries for 3D printing machines.
Purpose: The objective of this thesis is to explore smooth self-supporting topologies to obtain print-ready designs without needing post-processing methods for smoothing boundaries before fabrication and adding extra support structures during fabrication.
Approach: An element-based topology optimisation algorithm named Smooth-Edged Material Distribution for Optimising Topology (SEMDOT) is developed through introducing extra grid points to each element and using elemental volume fractions in Finite Element Analysis (FEA). In SEMDOT, multiple (dual) filtering steps are used instead of the single filtering step used in general element-based algorithms. The combination of SEMDOT and Langelaar's AM filter is used. Manufacturability experiments are set up in two typical AM technologies: Fused Deposition modelling (FDM) and Selective Laser Melting (SLM).
Findings: The proposed SEMDOT algorithm is capable of forming smooth topologies. A lower penalty coefficient can be used in SEMDOT, meaning that the optimisation problem is much closer to a convex problem. SEMDOT is capable of obtaining topological designs comparable or better than standard element-based algorithms. The use of multiple filtering steps enhances the flexibility of SEMDOT in exploring better performance and different topological designs. Incorporating Langelaar's AM filter enables SEMDOT to generate convergent self-supporting topologies which have been demonstrated to be printable using both FDM and SLM. Experimental results show that the conservative overhang angle criterion of 45 degrees would sacrifice more performance than is necessary to achieve the self-supporting goal. Topological designs obtained by SEMDOT can be directly manufactured by 3D printing machines without requiring redesign and post-processing.
Novelty: Accurate boundary information can be obtained using SEMDOT. Multiple filtering steps are used to obtain better performance. The Heaviside smooth function is used to implement the solid/void design of grid points and simultaneously mitigate numerical instabilities. The mathematical model of combining SEMDOT and Langelaar's AM filter is established.
Additive manufacturing (AM) possesses capability of building complicated parts that are otherwise difficult to manufacture by the conventional methods. However, the dimensional and geometric accuracies as well as surface quality of an AM-produced part are inferior to the conventionally machined part, which hinders the AM applications. Thus, an additive/subtractive hybrid manufacturing (ASHM) method is developed to take advantage of both the AM and precision subtractive manufacture (SM). However, the microstructures of the AMed parts are different from those of the conventional metallic parts. In addition, the residual stress induced by the AM stages influences the machined residual stress reconstruction in the subtractive stages. In order to investigate the effect of microstructure and the AM-induced residual stress on the machining characteristics, a milling experiment is conducted on AMed and wrought samples. The results of the cutting force, machined residual stress, and surface roughness are compared. It is found that the machining characteristics of AMed samples are different from those of wrought samples due to different microstructures and residual stress evolutions. The paper provides a guidance to the optimization of the processing parameters in the ASHM.