Author Archives: Boer Deng

Final Presentation

It is widely spread, attempted and explored, at the meanwhile, criticized, but undeniable that contemporary architecture as other fields of humanity discourses like media, entertainment, advertisement, as well as science discourses, is dominated by computers and computation paradigm. While the notion of computerization was first step, digital modeling is now certainly v into the era of computation in the process of calculation and reasoning of information. Contemporary designers are dealing with “genetic algorithms” as an approach of potential new interpretation of architecture. According to Manuel Delanda, demonstrated in his lecture of “Deleuze and the use of the genetic algorithm in architecture”, “genetic algorithm” is simply a set of simulations and rules in evolution of processing data in the computer. While in conventional design systems, there are various parameters such as topography, programs, facilities, structures, circulations and so on, which are supposed to be considered during the design process. The Algorithmic, and Parametric design is thinking method of transferring all these information and factors into algorithms to generate a logically or mathematical best solution for the design.

As we can see and experienced this whole semester of accomplish what we have in our crazy little brain, I found that design, manipulation and control of geometry of complex objects is very much well-developed and incredible accessible for all designers with different aspects. But unlike the theorist, and thinkers, the design by itself is not what practical architects and designers looking for. It is the building, or the actual space is what makes most of architects devote their life for. We now have witnessed and experienced some of this amazing technological driven approach of architecture design. The question is how practical this kind of design could be. Is it possible to build these complicated, curvy, blobby objects without costing ten times of the price, which only applicable for signature architects like, Frank Ghery, Zaha Hadid and so forth. And what is the social value, aesthetic value, humanity value of this type of architecture besides the exciting form. Is it true we can call this type of architecture adaptive? How about places with deep religious belief and rich historical value? Or, this system of thinking is supposed to adopt historical and religious parameters into the simulation, thus no conflict should be pop up? How about if we interpret this technology thing as a means to create an architecture that adapts to its occupation, on the one side it could mean that design is embodied within our experiences of spaces, palces, and the components of design are transparently embedded within the environment; on the other hand it could be depicted as which “society and space are combined as a unitary entity, brought together through a knowledge and understanding of computation as applied to our presence in the world.”(which reconnect me to our experience of parametric urbanism, where the preexisting factors and artificial manipulations, as well as the our new, proposed parameters are all unioned as one and being processed in the computer through the computation platforms like maya and grasshopper.

It becomes very important to consider all environmental proterties and material effects in design, to develop material systems which have the potential to mediate environment. From what I experienced during the process of produce the work with my stubborn mind of reasoning and programing, I also found that this genetic algorithmic think may start from generating forms with interesting looks. As it goes to the urbanism, the analysis factors including human behavior, economical value and development, larger view of the city political values started to involved into my process of generating the form and pattern. At the same time, all these factors have given me feedbacks I was expecting and they criticized themselves during the process as well.


P2_parametric city


P1_Boer Deng_Computer Simulation of Wind Load

For light weight membrane structures, the proof of safety under extreme windloads is one of the most
demanding tasks.

the load itself has stochastic nature and can only be described with very limited accuracy, when
the classical methods of civil engineering having deterministic nature are used. Assumptions
on wind profile, wind direction and strength of wind have to be made being only very limitedly
in accordance with reality.
Furthermore, the interaction of wind and structure has to be considered. In the most simple
case, only the direction of wind flow is influenced by a rigid construction. The pressure
distribution acts as a surface load on the structure. It is already more complicated when
the construction is deformed in such a way that the flow regime is modified which means that
an interaction of fluid and structure occurs. Most difficult is the prediction of a dynamic interaction.
Induced by the fluid flow, the stucture starts fluttering or galopping. This interaction
can lead to a complete failure of the structure.

A complete numerical computation of the interaction of wind and lightweight structures is
not yet used in civil engineering practice. One major reason is that already the formulation of
a realistic model leads to very significant difficulties. Furthermore, a reliable description of
the coupling phenomena is connected with a large number of mathematical and algorithmic
difficulties. Moreover, nearly all practically relevant cases are strictly three dimensional. A
reduction to two-dimensional flow around cross sections will lead only in very special cases
to realistic results. Yet, only the computation of a (typically turbulent) flow field around
large constructions is a very considerable challenge which can only be attacked by extreme
computational power.
Despite these general difficulties, significant progress in development and validation of
suitable coupling methods have been made during the past few years. Considering
this progress and assuming further increase of computational power gives rise to the
expectation that in only a few years numerical simulation will be a true alternative to wind
tunnel experiments in civil engineering practice. Therefore, the goal of this paper is to review
the most important algorithmic principles of fluid-structure interaction and to report on a detailed
comparison of wind tunnel experiment and numerical simulation. As an example a tent
construction is considered. In order to get a meaningful result, the wind tunnel model itself
was investigated by the numerical simulation and not the original “real life” construction.