exp.wiki NEU
WICHTIG: Das exp.wiki wurde aktualisiert und ist ab sofort unter einer neuen Adresse zu erreichen: https://exparch-www.uibk.ac.at/testwiki
Bitte das Anlegen neuer Seiten oder die Änderungen bestehender ab sofort ausschließlich dort durchführen.
Bestehende Projektseiten wurden/werden migriert und sind auch auf der neuen Plattform vorhanden.

Digital Aggregate

Aus exp.wiki

(Weitergeleitet von E7 Projekt Pfleger)
Wechseln zu: Navigation, Suche

Patrik Schumacher /Strukturelle morphologien

Betreuer: DI Markus Malin

Inhaltsverzeichnis

Digital Aggregate

by Martin Pfleger

Phase 1 /Physical-Digital Model

To better understand how aggregates reacted a model was needed. Inspired by Eiichi Matsua research I decided to experiment with the same aggregate particle. The particle is constructed by three match sticks forming a cross. It seemed like the natural choice considering the simplicity of its construction in order to produce a sufficient number of units in a fair amount of time. A second main factor in using this unit was that it had next to no interlocking qualities this insured that the stable state of the aggregate relied alone on the balance between friction gravity and weight of the unit.

My main objective was to inquire a better understanding how aggregates react when they are being poured when they settle and when they are in a fluid like state. […]



With the information gained I started to work on the digital model. For the simulation I used the crysis game engine from crytek. In order to compare the real with the digital aggregate I modeled the matchstick cross and formwork using the same proportions then exported it to the cry- engine. The next task was to adjust the surface friction and weight of the cross in the cry editor to match close to real conditions. The cross was then duplicated to match the same unit number compared to the real model.

The pouring process of the digital model was the one big difference to the real model. In the digital model it can be controlled in a way not possible in reality. In my first experiments I created a grid like unit arrangement suspended above the formwork, however after turning on gravity the outcome was rather troublesome once everything settled. Since the units at starting point had the same rotation and where arranged in a grid the result was a high accurace of regular stacking. I then started to use scripted unit configurations so units had multiple rotations and where arranged in a less strict grid leading to better results. […]



Since the results of the digital simulations with the matchstick unit where so pleasing I tested other unit shapes with interlocking qualities that could possibly be beneficial for stability. I also started to use a mix of different unit sizes to change the density of the model. […]




Phase 2 /Controllable Density

Since one can slightly control the outcome of the aggregate by arranging the starting conditions I wanted to create a method of controlling the areas of density throughout the aggregate. For this I scripted a 3d paint tool. The colors range from black to white each grade of color representing a certain unit size. The script creating the unit arrangement first reads the color grid and then defines the quantity and size of the units. […]


Phase 3 /Branching Void

My next focus was to have more control over the formwork creating the voids. In previous experiments I used balloons for the real model and spheres for the digital model. But in order to create path- like voids I was looking for a scripted approach to a branching systems and found the solution in L-Systems. After writing a generator for 2d Systems and testing common formulas I moved on to 3d systems trying to find ways to contain and direct branching systems in the aggregate formwork. […]




In addition to the branch generation I implemented metaballs (metaball plug-in Copyright (C) 2002 yuri@packeton.com) in my L system generator which would create the voids along the branching path. The metaballs are additionally scaled in size according to the density of the aggregates, the smaller the aggregate in the area the smaller the meatball void. […]



Combining Phase

As a final experiment I combined the Systems that where developed showing possible strengths and weaknesses of the systems and areas where Refinement is needed.

We have not seen the full potential of aggregates yet, but as technology evolves and simulation models become more precise our understanding of aggregate systems will grow.


Links

my aggregate simulation clip:

coming soon.

cry engin physiks:

http://de.youtube.com/watch?v=h3HZgTE7S1k

http://de.youtube.com/watch?v=XYEXhVsB0zw

http://de.youtube.com/watch?v=YG5qDeWHNmk

crytek:

http://www.crytek.com/technology/cryengine-2/specifications/

metaball plugin:

http://www.highend3d.com/maya/downloads/plugins/modeling/misc/2776.html

Morpho-Ecologies: Towards Heterogeneous Space In Architecture Design

http://www.amazon.com/Morpho-Ecologies-Towards-Heterogeneous-Architecture-Design/dp/190290253X/ref=sr_1_1?ie=UTF8&s=books&qid=1222631943&sr=1-1

Versatility and Vicissitude: Performance in Morpho-Ecological Design (Architectural Design)

http://www.amazon.com/Versatility-Vicissitude-Performance-Morpho-Ecological-Architectural/dp/0470516879


zurück zur Liste der Studentenprojekte

Attribute

Pfleger Martin Schumacher Patrik Budig Michael,Neumayr Robert,Malin Markus | E7 | E7 Projekt Pfleger | Strukturelle morphologien | 08SS | Bild:aggregate_mp_45.jpg | a

Fakten zu Digital AggregateRDF-Feed
BetreuerSchumacher Patrik  +, Budig Michael  +, Neumayr Robert  + und Malin Markus  +
ImageAggregate mp 45.jpg  +
LVE7  +
Semester08SS  +
ThemaStrukturelle morphologien  +
TitelE7 Projekt Pfleger  +
VerfasserPfleger Martin  +
Zuordnunga  +
Persönliche Werkzeuge