Interval Finite Elements as a Basis for Generalized Models of Uncertainty in Structural Analysis

UVS
13. Juni 2007

Im Rahmen der SFB-Seminare des Sonderforschungsbereichs 528 „Textile Bewehrungen zur bautechnischen Verstärkung und Instandsetzung“ spricht am 19. Juni 2007 Prof. Ph. D. Rafi L. Muhanna von der Georgia Tech, Atlanta, über „Interval Finite Elements as a Basis for Generalized Models of Uncertainty in Structural Analysis„. Gäste sind zu dem Vortrag, der im Sitzungszimmer Raum 67 im Beyer-Bau um 16:30 Uhr stattfindet, herzlich willkommen.

Aus dem Abstract:

In recent years, there has been an increased interest in the modeling and analysis of engineered systems under uncertainties. The interest stems from the fact that numerous sources of uncertainty exist in reality and arise in the modeling and analysis process. Some of these uncertainties stem from factors that are inherently random (or aleatory). Others arise from the lack of knowledge (or epistemic).

These facts make us to believe that uncertainty is unavoidable in engineering systems. Solid and structural mechanics entails uncertainties in material, geometry and load parameters.

Latest scientific and engineering advances have started to recognize the need for defining multiple types of uncertainty. Probabilistic approach is the traditional approach that requires sufficient information to validate the probabilistic model. Probabilistic modeling cannot handle situations with incomplete or little information on which to evaluate a probability, or when that information is nonspecific, ambiguous, or conflicting. Many interval-based uncertainty models have been developed to treat such situations.

This talk presents an interval approach for the treatment of parameter uncertainty for static structural mechanics problems. Uncertain parameters are introduced in the form of unknown but bounded quantities (intervals). Interval analysis is applied to the Finite Element Method (FEM) to analyze the system response due to uncertain stiffness and loading.

Print Friendly, PDF & Email