Refine
Has Fulltext
- yes (6) (remove)
Document Type
- Conference Proceeding (3)
- Article (2)
- Doctoral Thesis (1)
Institute
Keywords
- Optimierung (6) (remove)
Year of publication
- 2004 (6) (remove)
Recently, many reseraches on active control systems of building structures are preformed based on modern control theory and are installed real buildings. The authors have already proposed intelligent fuzzy optimal active control (IFOAC) systems. IFOAC systems imitate intelligent activities of human brains such as prediction, adaptation, decision-kaking and so on. In IFOAC systems, objective and subjective judgements on the active control can be taken into account. However, IFOAC systems are considered to be suitable for far-field erathquake and control effect becomes small in case of near-field earthqaukes which include a few velosity pules with large amplitudes. To improve control effect in case of near-souece earthquakes, the authors have also proposed hybrid control (HC) systems, in which IFOAC systems and fuzzy control system are combined. In HC systems, the fuzzy control systems are introduced as a reflective fuzzy active control (RFAC) system and imitates spinal reflection of human. In HC systems, active control forces are activated to buildings in accordance with switching rules on active control forces. In this paper, optimizations on fuzzy control rules in RFAC system and switching rules of active control forces in HC system are performed by Parameter-Free Genetic Algorithms (PfGAs). Here, the optimization is performed by using different earthquake inputs. The results of digital simulations show that the HC system can reduce maximal response displacements under restrictions on strokes of the actuator effectively in case of a near-source earthquake and the effectiveness of the proposed HC system is discussed and clarified.
The paper gives a general overview and concerns with a specified set of computer-aided analysis modules for hybrid structures loaded by extreme excitations. All problems are solved by methods of linear, quadratic or nonlinear mathematical optimization, that leads to very effective and economic design solutions. All approaches are derived from general optimization problem that can be easily altered to conform to specific design tasks. Some advantages and possibilities of hybrid structural modeling (single or mixed model-supported) are discussed. The methods will be illustrated by an example structure and optimization schemes.
This paper presents results of a study on distributed, or parallel, evolutionary computation in the topological design of steel structural systems in tall buildings. It describes results of extensive experimental research on various parallel evolutionary architectures applied to a complex structural design problem. The experiments were conducted using Inventor 2003, a networkbased evolutionary design support tool developed at George Mason University. First, a general introduction to evolutionary computation is provided with an emphasis on recent developments in parallel evolutionary architectures. Next, a discussion of conceptual design of steel structural systems in tall buildings is presented. Further, Inventor 2003 is briefly introduced as well as its design representation and evolutionary computation characteristics. Next, the results obtained from systematic design experiments conducted with Inventor 2003 are discussed. The objective of these experiments was to qualitatively and quantitatively investigate evolution of steel structural systems in tall buildings during a distributed evolutionary design process as well as to compare efficiency and effectiveness of various parallel evolutionary architectures with the traditional evolutionary design approaches. Two connectivity topologies (ring topology and fully-connected topology) have been investigated for four populations of structural designs evolving in parallel and using various migration strategies. Also, results of the initial sensitivity studies are reported in which two ways of initializing distributed evolutionary design processes were investigated, using either arbitrarily selected designs as initial parents or randomly generated ones. Finally, initial research conclusions are presented.
We present a software prototype for fluid flow problems in civil engineering, which combines essential features of Computational Steering approaches with efficient methods for model transfer and high performance computing. The main components of the system are described: - The modeler with a focus on the data management of the product model - The pre-processing and the post-processing toolkit - The simulation kernel based on the Lattice Boltzmann method - The required hardware for real-time computing
Current software solutions for real estate planning, construction and use, do not model the complete life cycle of a building. Well-integrated software tools exist for the planning and construction phases. Data integrity exists throughout the planning and construction phases, but problems occur at the transition to the use-phase. At this interface, the complete data set of planning and execution gets lost. Another software deficiency is that current software solutions don’t handle construction work and maintenance work equally. This is why a new software generation is demanded, which continuously covers the entire workflow process from the planning phase to the demolition of a building. New data concepts have to be developed, which allow bringing work items for construction together with work items for real estate use.
Das Ziel der vorliegenden Arbeit besteht in der Entwicklung einer Strategie zur physikalisch nichtlinearen Analyse von Aussteifungssystemen. Der Anwendungsschwerpunkt umfasst neben dem traditionellen Aufgabenumfang zur Analyse neu zu errichtender Tragwerke gleichzeitig auch Planungsaufgaben, die mit Umbau- und Sanierungsmaßnahmen verbunden sind. Veränderungen, die sich während der Nutzungsgeschichte oder im Revitalisierungsprozess ergeben, werden in den Berechnungsmodellen berücksichtigt. In vielen Fällen ist es aus planerischer Sicht zweckmäßig, die Nichtlinearität des Materialverhaltens zur Erschließung von Tragreserven in den normativen Nachweiskonzepten mit einzubeziehen. Der damit verbundene numerische Aufwand wird durch die Verwendung separater Modelle zur Erfassung des Querschnitts- und des Systemtragverhaltens begrenzt, ohne die Komplexität der Aufgabenstellung zu reduzieren. Aus detaillierten Querschnittsuntersuchungen der Tragwände werden integrale Materialbeziehungen abgeleitet, welche die Grundlage für die nichtlineare Tragwerksanalyse darstellen. Die Modellbildung gegliederter Aussteifungswände basiert auf deren Zerlegung in ebene finite Stabsegmente, die sich durch die Diskretisierung in Längs- und in Querrichtung ergeben. Zusätzlich zu den an den Stabenden angreifenden Normalkräften, Querkräften und Biegemomenten werden an den Elementlängsrändern Schubbeanspruchungen erfasst. Die physikalische Nichtlinearität wird durch die Einbeziehung integraler Materialbeziehungen an den Segmenträndern berücksichtigt. Die numerische Umsetzung erfolgt mit Methoden der mathematischen Optimierung. Die Leistungsfähigkeit der Berechnungsstrategie wird exemplarisch anhand von Untersuchungen an Aussteifungssystemen in Großtafelbauweise nachgewiesen.