TUD » CRC 528 » Program » Project Field E » Subproject E3
Logo CRC528
Logo CRC528

Subproject E3

Safety assessment of textile strengthened structures

Director  Staff  Goals  Methods  Results  Publications  Diplom thesises

Director

4th Founding periode

Dr.-Ing. Jan-Uwe Sickert
Institute for Structural Analysis

1st - 3rd Founding period

Prof. Dr.-Ing. habil. Bernd Möller; leading
Prof. Dr.-Ing. Wolfgang Graf
Institute for Structural Analysis

Staff

Prof. Dr.-Ing. Wolfgang Graf (advisory)
Prof. Dr.-Ing. habil. Michael Kaliske (advisory)
Prof. Dr.-Ing. habil. Bernd Möller (advisory)
Dipl.-Ing. Christian Jenkel

Goals

Fig. 1: Probability density function
with fuzzy limit state function and
multi modal importance dansity function
ot the FAIS algorithm.
The reliability of structures depends on their lifetime and a multitude of single effects such as material damage, mechanical damage, corrosion, number and quality of maintenance. It is time-dependent. Therefore, the safety assessment of existing structures is always the evaluation of a specific case for which in most cases the number of observations is insufficiently large for a statistical evaluation free of doubt.

The development of models and algorithms to describe the uncertainty concerning the analysis of textile-strengthening structures in a different way started in the 1st working period. The uncertainty was modelled by either fuzziness or fuzzy randomness. It has been found, that fuzzy randomness constitutes an essentially extensive basis compared with fuzziness. Fuzzy randomness includes both randomness and fuzziness as special cases and provides the consideration of different types of uncertainty (randomness, fuzziness and fuzzy randomness) simultaneously.

Therefore, in the 2nd working period the theory to fuzzy randomness is expanded and the safety concept 1 based on it is developed further and extended, e.g with the introduction of fuzzy random functions in order to describe spatial and time dependencies of uncertain fluctuating input variables.

On the basis of the fuzzy probabilistic safety concept, the to safety assessment inverse task will be solved in the 3rd working period, i.e. design of the textil reinforced strengthening. The set of permissible solutions for the safety goal oriented design is evaluated by the aid of the robustness measure. Each single uncertain input parameter is assessed according to structural reserves and sensitivity. This algorithm supports the choice of a deterministic design variant.

Top

Methods

The basis for the development of the safety concepts are the uncertain data analysis, the fuzzy structural analysis, the fuzzy probabilistic safety assessment, the possibility theory based safety assessment and the compilation of structural damage. The application of the concepts for the safety evaluation of textile-strengthened buildings requires the integration of the computation model of the subproject D2 "Numerical simulation of the load-bearing behaviour of textile strengthened constructions". The material parameters, geometry, and load variables of existing structures and textile-reinforced strengthening layers possess also data uncertainty, which cannot be described by stochastic models satisfyingly. This in particular informal data uncertainty can be better modelled with fuzzy randomness or fuzziness.

Uncertainty models

Uncertain data analysis

Sructural analysis with uncertain parameters

Safety assessment

Time-dependent fuzy random data uncertainty leads to uncertain time variant reliability. The reliablity of textil strengthened structures can computed under consideration of the available data with three different safety concepts.


Fig. 2: Fuzzy probabilistic time-dependent limite state concept.

Top

Results

The methods and algorithms specified above are transferred into software, which are used for the solution of the example problems.

Uncertain data analysis

Structural analysis with uncertain parameters

Possibility theory based safety assessment

Fuzzy probabilistic safety assessment

Top

Publications

Year 2005, 2004, 2003, 2002, 2001, 2000, 1999, 1998

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

Seitenanfang

Diplomarbeiten

Seitenanfang