Article

Modularity and flexibility at the small scale: evolving continuous material variation with stereolithography

Authors:
Article

Modularity and flexibility at the small scale: evolving continuous material variation with stereolithography

If you want to read the PDF, try requesting it from the authors.

Abstract

In this paper, we introduce a technique by which the internal material properties of an object can be optimised at a microstructural level (5x10-5m) to counteract the forces that are applied to it. These can then be fabricated using the rapid prototyping method of stereolithography. The proposed technique is analogous to principles of mass customization and takes advantage of a flexible module to create complex structures in a manner that is computationally efficient and effective. The process is two-staged, in which a genetic algorithm evolves the topology of the microstructure and a second algorithm incorporating Finite Element Analysis then optimises the geometry. The examples shown are designed specifically for the fabrication technique, but the method and general principles are applicable to structural problems at any scale.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Bendsøe and Sigmund split the problem into the two coupled sub-problems of local anisotropy, and material distribution, which can be addressed in the problem representation used here by altering the direction and thickness of the struts respectively. Previous work by the author (Hanna and Haroun Mahdavi, 2004) has resulted in a method for rapid optimisation of large and complex structures using this modular 'unit cube' approach. By confining optimisation to local unit areas, this prior method finds fit structures with a member count too high to manage using a standard global optimisation, at greatly improved speed. ...
... The design problem is a simple cantilever like the examples shown inFigure 4, for which modular gradient descent has been found effective in increasing stiffness (Hanna and Haroun Mahdavi, 2004). The overall distribution of loads throughout the object was first determined using FEM on a regular grid of 8- node isoparametric volumes to estimate the local relative displacements in each axis for a homogeneous material, and these then taken as the local loads x with which to optimise the structure of each unit cube. ...
Article
While structural optimisation is usually handled by iterative methods requiring repeated samples of a physics-based model, this process can be computationally demanding. Given a set of previously optimised structures of the same topology, this paper uses inductive learning to replace this optimisation process entirely by deriving a function that directly maps any given load to an optimal geometry. A support vector machine is trained to determine the optimal geometry of individual modules of a space frame structure given a specified load condition. Structures produced by learning are compared against those found by a standard gradient descent optimisation, both as individual modules and then as a composite structure. The primary motivation for this is speed, and results show the process is highly efficient for cases in which similar optimisations must be performed repeatedly. The function learned by the algorithm can approximate the result of optimisation very closely after sufficient training, and has also been found effective at generalising the underlying optima to produce structures that perform better than those found by standard iterative methods.
... Within the field architecture Correa et al. [5] engage in 3D printing in order to develop aim to develop custom wood grain structures with unique anisotropic and hygroscopic properties. Similarly Hanna [6], Oxman [7], Gardiner [8] and Palz [9] have demonstrated the potential to design and fabricate graded materials using computational design strategies and 3D printing technology. ...
Conference Paper
Full-text available
Additive manufacturing technologies makes it possible to control the target deformability behavior of global geometry by varying the material deposition of microstructure. This offers new ways to fabricate cellular materials with differentiated material behaviors that go beyond mimicking the structure and mechanical properties of materials found in nature. However, material limitations and a lack of appropriate design tools and models have confined the usefulness of 3D printed cellular materials. This paper introduces and describes a novel design model that allows to analyze and exploit microstructures in the context of architecture. Specifically, this paper discusses the use of 3D cellular microstructures to fabricate flat deformable panels with predefined elastic behavior that can be used in architecture to approximate arbitrarily complex shapes. In the first part this paper will introduce and describe the design tool, which was developed by an interdisciplinary research team, comprising researchers from architecture, computer graphics and civil engineering. The model takes into account parameters affecting the structural response of 3D cellular microstructures, such as the volume fraction, orientation distribution, aspect ratio within each cell as well as the cell geometry and relative density are included in the model. To guide the design of cellular microstructures, we developed an integrated, parameterized, multi-scale model. Computational homogenization based on finite element is employed to obtain the elastic properties of the material at multiple scales. In the second part of this paper, we will discuss the findings of a case-study undertaken by the authors in which the previously described approach was employed and tested.
... An advantage of such manufacturing technologies as laser sintering and stereolithography is that they can manipulate geometry at a scale fine enough to create internal meso-structures of large objects, effectively customising the material properties of designed objects (Figure 8.6). Density of materials can be modified while optimising for stiffness under specific loads (Hanna and Haroun Mahdavi, 2004) as can properties such as Poisson's ratio that have a profound effect on compliant behaviour. The optimisation of millions of individual members is required, along with the necessary structural simulation, but this can be costly and time-consuming. ...
Book
The relationship between representation and the represented is examined here through the notion of persistent modelling. This notion is not novel to the activity of architectural design if it is considered as describing a continued active and iterative engagement with design concerns – an evident characteristic of architectural practice. But the persistence in persistent modelling can also be understood to apply in other ways, reflecting and anticipating extended roles for representation. This book identifies three principle areas in which these extensions are becoming apparent within contemporary practice: the duration of active influence that representation can hold in relation to the represented; the means, methods and media through which representations are constructed and used; and what it is that is being represented. Featuring contributions from some of the world’s most advanced thinkers on this subject, this book makes essential reading for anyone considering new ways of thinking about architecture. In drawing upon both historical and contemporary perspectives this book provides evidence of the ways in which relations between representation and the represented continue to be reconsidered. It also provides critical insight into the use of contemporary modelling tools and methods, together with an examination of the implications their use has within the territories of architectural design, realisation and experience.
... Case study 5.3. Sarat Babu's "Microkinetics" (see Figure 8) explores micro-level structures 24,25 embedded in a matrix that produces physical characteristics outside of the expectations of their three-dimensional forms. A combination of selective laser addition (SLA) rapid prototyping and cast polyurethane and silicon allowed the creation of an "alphabet" of micro functions, as illustrated in Figure 8. ...
Article
Full-text available
Experimentation is often considered a constituent part of the design process and designing in general, yet its exact function and identity is open to a wide variety of interpretations. For example experimentation is often confused with differentiation or iteration. The relationship between scientific and industrial design experimentation; rationale, process, objectives and methods have rarely been considered. This paper will reflect on the role of experimentation in industrial design and compare its activity to that in the scientific world. Through case studies (from the new Experimental design strand at the Royal College of Art and Imperial College’s Innovation Design Engineering dual MA/MSc) a methodology of balanced mapping and exploration will be discussed. Scientific experiments have to be repeatable in order to be valid, yet in the design world this is often impossible due to the tackling of ‘wicked’ problems that change the very nature of the problem itself, preventing repetition. In practical terms designers value a unique ‘one-off’ approach helping to guarantee the innovation and originality of their solution. At the heart of this enquiry is the difference between design experimentation: designing using experimental methods and experimental design, a fundamental creative methodology for the foundation of new industrial designs, systems and technologies.
... Given a parameterized structure and a set of loading conditions, it has been shown that various optimization algorithms can be used to design an effective shape to counter the given load. Search procedures including gradient descent (GD) and genetic algorithms (GA) make repeated evaluations of the strength of different structures to do this (Schoenhauer 1996;Von Buelow 2002;Chen 2002;Hanna and Mahdavi 2004). If the load conditions change, the optimal structure will also be different and the optimization can be rerun to find the new shape. ...
Chapter
Full-text available
A support vector machine is trained to produce optimal structures. The problem of structural optimisation is typically solved by a search procedure that samples and repeatedly evaluates a physicsbased model, but this process is computationally demanding. Instead, the use of a learning algorithm to generate new structures based on previously optimized examples is described that provides enormous computational saving. The results have shown that the predicted structures are accurate, and the process is highly efficient for cases in which similar optimisations must be performed repeatedly, especially as the number of such optimisations grows.
Chapter
Full-text available
In response to global challenges of resource scarcity, increasing attention is being paid to bio-based materials - a domain that covers familiar materials such as timber and emerging materials such as bio-plastics and mycelium composites. The ability to observe, analyse, simulate, and design with their interior heterogeneity and behaviour over time is a necessity for a bio-based and cyclical material practice and opens a deep reservoir of creative and technical innovation potentials within architecture and aligned design practices.This paper describes a research inquiry which seeks to integrate volumetric material data acquired through non-intrusive methods into materially-led digital design workflows. The inquiry is developed as a set of computational tools and approaches to architectural modelling, and demonstrated through three main material tracks: structural glue-laminated timber assemblies, mycelium composites, and bio-luminescent bacteria substrates. Each addresses the acquisition, analysis, and simulation of deep volumetric material data at different scales and in different deployment contexts. In doing so, we demonstrate a novel shift in the digital modelling of bio-based architectural materials and set out its implications for new design practices that deeply embed the individuality and temporality of materials.We contribute a perspective on the possibilities afforded by a volumetric modelling approach to bio-architecture and a computational framework for operating with volume data of heterogeneous materials.
Article
Full-text available
Additive manufacturing (AM) processes enable the production of functional parts with complex geometries, multi-materials, as well as individualized mass production. Another significant benefit of AM is the ability to produce optimized geometries with near perfect strength to weight ratios. Weight plays a crucial role in many functional parts such as parts used in automotive and aeronautic industries. Current topology optimization techniques, a powerful tool for weight reduction and product optimization in general, do not work well for such kind of process since AM methods necessitate applying own dedicated design rules. This paper reports a product/process optimization study of a simple test case geometry (C-Clip), where structural optimization has been applied using an innovative approach based on the design of lattice structure feasible thanks to additive process adoption. Moreover, it has been conducted a study to evaluate the possible advantages offered by the integration of the two previous approaches in order to verify the required design specifications. The aim of the work has been to evaluate the potentiality offered by the integration of the two structural optimization approaches (topological and lattice structures design) to obtain innovative and highly performing structures. This activity represents a necessary step for the definition and the subsequent development of a methodology aimed to the creation of structures obtained with this combined design approach. In order to define an objective evaluation of the component performances, appropriate Key Performance Indicators (KPI) have been developed. An engineering intelligence tool has been used to post process the generated optimization results for the three different approaches. Finally, the first three “best” structural solutions have been manufactured by 3D printing machine, with scaled dimensions, in order to evaluate the printing time considering the geometry complexity for the chosen structural layout in order to have useful feedbacks on Product/Process choices interaction.
Thesis
Full-text available
This thesis examines the activity of designing and making across geographical distances. Through critical reflection on the author’s own design practice, and the initiation of new collaborative design projects, the thesis develops and tests the concept of translocated making. Translocated making is explored as a form of production where activity, influence, media and understanding are shared across multiple locations, both physically and virtually. The author’s own design practice is situated and analysed in the context of how the manufacturing possibilities of mature western markets have developed since the 1990s. With the rise of digital technologies, greater degrees of interaction and collaboration between designer and manufacturer have been enabled; however, it is found that the effects of geographical and cultural distance on designing and making have not been systematically researched, nor understood from the perspective of the designer as creative agent. Drawing on theories of cultural interaction, and in particular Appadurai’s theory of suffixscapes, the thesis sets out a framework for experimenting with and analysing the effects of cultural and geographical difference on designing and making. Collaboration across distance is established as a key mode of designing and making, bringing to light the effects of difference in design outcomes. Gujarat in India is the location for a series of design projects in urban and remote locations. These are investigated through designing research and researching through design methods. New knowledge is articulated through the way in which the design projects allow for the testing and reflection upon theories of cultural interaction. The experimental design methods employed in the projects show how exchanging differences through collaboration in digital and analogue media can create new artefacts with hybrid cultural values. The value of this research for designers and makers in advanced and developing economies is through helping them to understand the possibilities of collaboration across geographical distance. For academics and researchers the value lies in critiquing and further developing practice-based design research methods, and in exploring the longer term strategic, creative and cultural changes that globalisation and digital technologies are bringing to all forms of design and manufacture.
Conference Paper
Full-text available
This paper presents the development of an SLS 3D-printed auxetic structure actuated to a predefined form by an embedded pneumatic network through an iterative process of feedback between digital simulation and physical testing. This feedback process is critical to the development of a more accurate predictive model and to compose the geometry of the suggested structure. An approach based on the emergence of final structure from the convergence of the behavior of sub--structures has been introduced from the beginning and a methodology based on the analysis and synthesis of the simplest sub-system is the core of this research. The results indicate a promising simulation environment and a novel methodology for the design and fabrication of auxetic structures with embedded pneumatic actuation. This exploratory research suggests a fertile research space within the field of adaptive architecture and kinetic design.
Article
Architecture, the art and science of designing the built environment, has proved to be a highly challenging area of application for computer science. This may be partly to do with the difficulty experienced by those without an architectural training in trying to understand exactly what design of the built environment entails. This is hardly helped by the descriptions that architects themselves give of this activity which tend either to paint it as more rational than it is, or to surround it with the mystique of the individual artistic genius. However, computers are now firmly placed at the centre of architectural practice, and the number of computers embedded in buildings themselves is increasing rapidly. This paper reviews the relationship between computing and architecture by describing a series of recent and current research projects in which the boundary between the digital and real world is explored, taking as their basis architectural theories derived from space syntax research. It describes the changing role of computing as a design medium, the embedding of computing in the buildings we design, and the broader theoretical contribution that architectural research is beginning to make to our understanding of the nature of human cognition and so to the field of artificial intelligence. It concludes that the relationship between the discipline of architecture and contemporary developments in the theory surrounding computing result from the need for both to be based on an epistemology of the extended mind.
Article
A support vector machine is trained to produce optimal structures. The problem of structural optimization is typically solved by a search procedure that samples and repeatedly evaluates a physics-based model, but this process is computationally demanding. Instead, the use of a learning algorithm to generate new structures based on previously optimized examples is described that provides enormous computational saving. The results have shown that the predicted structures are accurate, and the process is highly efficient for cases in which similar optimizations must be performed repeatedly, especially as the number of such optimisations grows.
ResearchGate has not been able to resolve any references for this publication.