Workshops and Master Classes

Workshops and master classes will be held on Saturday, July 14th and/or Sunday July 15th, from approximately 9am-5pm daily with a break for lunch (not included).

Registration is limited and is accessible through the sympsium’s main registration page. Costs are $100 for 2 day courses and $50 for 1 day courses, and participants must also be registered for the IASS 2018 symposium. 

Participants are expected to bring their own laptops. For specific workshop or master class questions, please email the course organizers. 

 

Computational Graphic Statics using COMPAS
2 Day Master Class (July 14-15)

by

Prof. Philippe Block
Dr. Tom Van Mele, van.mele@arch.ethz.ch
Dr. Tomás Méndez Echenagucia
Patrick Ohlbrock
Juney Lee

 

In this master class participants will learn how to use advanced 2D and 3D graphic statics, using computational methods through COMPAS, an open source, Python-based framework for “computational research in architecture and structures” developed by the Block Research Group at ETH Zurich. On the first day, we will set up a development environment for scientific programming that will be used throughout the workshop, followed by a “scientific Python 101”. The introduction covers the key concepts of the COMPAS framework and a series of representative code examples related to computational graphic statics. The second day we will explore the structural design possibilities provided by specialised COMPAS packages for working with Algebraic Graph Statics (compas_ags), Thrust Network Analysis (compas_tna) and 3D Graphic Statics (compas_3gs). 

 

 

 

Machine Learning for Structural Design Space Exploration
2 Day Master Class (July 14-15)

b

Thornton Tomasetti COREStudio
Dan Reynolds, DReynolds@ThorntonTomasetti.com
Margaret Wang, MWang@ThorntonTomasetti.com
Kam-Ming Mark Tam, MTam@ThorntonTomasetti.com 

 

 

Design exploration of architectural structures is a multi-dimensional non-linear process that involves many iterations and feedback from both designers and engineers.  However, the process for obtaining performance feedback, which require analyses and simulation, can be time-consuming and computationally intensive—thereby limiting the capacity for designers to effectively and expediently explore a design space.  To aid in this process, designers and engineers have begun adopting various data-driven techniques, such as Machine Learning (ML), to develop prediction models that enable designers and engineers to obtain quick and reliable approximate measures of performance without the need to run an analysis and validation on each design option. Participants will learn about the development of prediction models using ML algorithms and apply the models to their own geometric massings.   

 

 

Optimization-Driven Creativity in Structural Design
1 Day Workshop (July 15)

by

Matthew Gilbert, University of Sheffield, m.gilbert@sheffield.ac.uk
Linwei He, University of Sheffield
Paul Shepherd, University of Bath
Catherine Rankine, Arup
Helen Fairclough, University of Sheffield
Andy Tyas, University of Sheffield
Antiopi Koronaki, University of Bath

Layout Optimization is a powerful technique that has to date been little exploited by the IASS community, or by construction professionals more generally. The technique can identify efficient structural forms for a range of constructions, including long-span roofs, gridshells, space-frames and tall building superstructures. It has some similarities with the more familiar topology optimization, but is better suited to the kinds of skeletal structures encountered in building structures.

Current UK government-funded research at the Universities of Sheffield, Bath and Edinburgh seeks to address a range of barriers to uptake. Outcomes thus far are promising and we have developed bespoke software called “FORM” (both as a stand-alone application and as a plug-in for Grasshopper) which makes it easy for a designer to use these tools at the concept design stage. We wish to share the methods developed with members of the IASS community at this workshop, while also providing background on the theory and hands-on workshop sessions using the software.

Participants will benefit by being exposed to state-of-the-art methods which are likely to be unfamiliar to them and will challenge their preconceptions about the value and power of “optimization”, particularly as a creative design tool.

 

 

 

 

 

Structural Morphology:  Interactive Exploration of Transdisciplinary Research Challenges
2 Day Master Class (July 14-15) 

by

Niels De Temmerman, Vrije Universiteit Brussel, Belgium, niels.de.temmerman@vub.be
Tomohiro Tachi, University of Tokyo, Japan
Rupert Malecek, UIBK, Austria
Christoph Gengnagel, UDK-Berlin, Germany
Julian Lienhard, TUM, Germany
Philippe Block, ETH Zürich, Switzerland
Corentin Fivet, EPFL, Switzerland
 Gennaro Senatore, EPFL, Switzerland

As the SMG group is a working group with a large number of members (70+) and is built around four study groups with a specific topic of exploration, an intensified interaction is one of the focal points of its current activities.

Therefore, this master class will take the form of several combined master classes, each dealing with a specific topic, based on research questions that have surfaced within the study groups that make up the Structural Morphology Group. Delving deeper into these subtopics will not only bring hand-on solutions and interesting insights that can act as a catalyst for further research and exploration, they also will facilitate interaction between current and new study group members. Additionally (and potentially), this could lead to a new field of study to be pursued within SMG. 

Currently the following cross-over topics are proposed, but additional themes can be put forward:

–        Transformables + active bending
–        Active bending + origami
–        Scissor structures and surface folding
–        Transformables and graphic statics

The output of these combined masterclasses, will be presented in the SMG Common Session at the conference, to show future potential. 

 

Structural NURBS – Form-Finding and Design of Complex Structures with Isogeometric Methods 
2 Day Master Class (July 14-15)

by

Philipp Längst, M.Sc., str.ucture GmbH, laengst@str-ucture.com
Anna Bauer, M.Sc., Chair of Structural Analysis Structural Analysis, Technical University of Munich, am.bauer@tum.de
Riccardo La Magna, Dr.-Ing., str.ucture GmbH, lamagna@str-ucture.com

The master class will explore mesh-free modelling and simulation of complex structural systems with the Isogeometric Analysis method. Isogeometric Analysis (IGA) is a new and innovative computational approach within Finite Element Analysis, which allows users to perform simulations of structural behaviour directly on the NURBS parametrization of CAD geometry objects. The master class will use Kiwi3d, a new IGA-plugin developed by the master class organizers for Rhino and Grasshopper. Kiwi3d integrates modules for geometrical linear and nonlinear analysis as well as cutting-edge form-finding algorithms and building process routines directly embedded in the native NURBS definition of the model.

The master class aims to provide participants with fundamental knowledge in Isogeometric Analysis and the core principles for the correct setup of simulation routines. The course will consist of three parts. In the first part, the participants will be introduced to the principles and the computational method of IGA. These will be demonstrated through a variety of examples and exercises. The second part will focus on developing complex models and simulations within the fully integrated NURBS-based design-to-analysis pipeline. This will also include the form-finding and analysis of membrane and bending active structures, modelling and simulation of building processes and optimization-driven design. In the third part of the master class, the participants will develop personal design ideas with the support of the tutors. Optionally, small scale physical models will also be realized. All design and simulation results will be presented and discussed within the group.

http://lado.edu/library/?page_id=505