Evolutionary Computational Design: For Structural Optimization

AI Creative Challenge 4.0_ Winner01

Become A Digital Member

Subscribe only for €3.99 per month.
Cancel anytime!

Advance your design skills

join PAACADEMY’s online workshops to learn more about parametric and computational design

Evolutionary Computational Design

Introduction to the Workshop:

Evolutionary Computation (EC) is a sub-field of artificial intelligence (AI) which used the “Evolutionary Algorithm” (EAs) to solve problems that have too many variables for traditional algorithms. This method uses to select the best alternatives between various solutions based on problems’ criteria and constraints known as “Optimization”. The engineering community (including architects and structural Eng.) could benefit from this design method as a rapid and high-quality decision-making method and tool to increase the speed and quality of their design choice in the building design field.

This workshop’s participants would learn how to use the EAs to study the effect of architectural design decisions on buildings’ structures’ performance and how to select optimum design alternatives among various design solutions. To achieve this goal, we use “Karamba3D” as a Finite Element (FE) analysis tool which provides an accurate analysis of spatial trusses, frames, and shells in the early design stage, and, “Galapagos” as an optimization tool based on the genetic algorithm. They are both fully embedded in the parametric design environment of Grasshopper, a plugin for the 3D modeling tool Rhino. This makes it easy to combine parameterized geometric models, finite element calculations, and evolutionary optimization algorithms altogether.

Scope of the Workshop:

This workshop is for architecture and structural designers and/or students who are interested in the simultaneous design of structure and architecture. All modeling and computational analysis will be done in the grasshopper/Rhino7. Having beginner to intermediate knowledge of Grasshopper is a must. Karamba3D will be used for modeling and analyzing the structure during sessions. No prior Karamba3D knowledge is required. The trial Karamba3D version is sufficient. The Evolutionary Optimization Algorithms will be explained from scratch and Galapagos, as the single objective optimization tool based on the genetic algorithm will be introduced. Prior knowledge of optimization algorithms is an advantage but not necessary.

Methodology:

This workshop presents a computer-aided design framework for the generation of non-standard structural forms in static equilibrium using Karamba3D and Galapagos, as Finite Element analysis tools and Genetic Algorithm optimization tools, respectively. The design workflow relies on the implementation of a series of operations (generation, evaluation, optimization, and regeneration) that allow to the creation of multiple design alternatives and to navigate in the architectural design space according to objective and subjective Criteria defined by the designer.

Objectives:

Participants of this workshop will:

  • Understand the main criteria and evaluation factors of a “FE analysis”,
  • Learn how to model and analyze simple structures using “Karamba3D”,
  • Understand the basic concepts and application of “Evolutionary Algorithms”,
  • Learn how to use a “Genetic Algorithm” for optimization and design framework in the architectural design process, and Learn how to use “Galapagos” to define a generative structure and architecture optimization workflow.

Program:

First Session: June 3rd

  • Introduction to Karamba3D
  • Setup a structural statical model
  • Creating structural models and analysis scenario
  • “Evolutionary Algorithms” and “Genetic Algorithms” are basic concepts

Second Session: June 4th

  • Assignment 01: Canopy Design
  • Structural performance optimization using Galapagos

Note: the participant can either design their own canopy or work on a simple algorithm prior prepared by the instructor. Note that while we are using the trial version of Karamba3D the canopy elements must be restricted to 20 beams & 50 shells.

Evolutionary Computational Design
©John Gollings, MPavilion, Melbourne

Software:

Important Notes:

Instructor:

Ghazal Javidannia

Evolutionary Computational Design - Ghazal Javidannia

Ghazal Javidannia has a Ph.D. from TMU University of Tehran, Iran. Her focus is on “Computational Design & Optimization using “Evolutionary Algorithms” and her expertise is on “Generative and Interactive” architectural design based on buildings’ structural performance. She has published multiple papers in peer-review Journals and has presented at various conferences including eCAADe and SimAUD. She also has held multiple national and international workshops in the field of architecture and structure Simulants Design” using “Evolutionary Algorithms”. Now, she collaborates with major design and building companies on building performance simulation and analysis in the early stages of design.

Evolutionary Computational Design
Share with a friend:
Courses:

Learn about parametric and computational from the online courses at the PAACADEMY:

Leave a Comment

Your email address will not be published. Required fields are marked *

Become A Digital Member

Subscribe only for €3.99 per month. Cancel anytime!

Weekly Newsletter in Your Inbox

Explore More

Sponsored Content

Subscribe to our weekly newsletter