|
|
Inverse reliability analysis of prestressed bridge
Lipowczan, Martin ; Novák, Drahomír (referee) ; Lehký, David (advisor)
The proposed diploma thesis deals with the application of methodology and tools of inverse analysis for design of selected structural parameters using a fully probabilistic analysis to determine the level of its reliability. The method based on artificial neural networks is used to approximate the inverse function. The inverse analysis was carried out in two ways that differs in the method of obtaining reliability indicators. The structure analyzed in this work was an existing bridge. The year of construction is estimated approximately between the years 1955 to 1960. The bridge is located close to the Uherský Ostroh. It is a one-piece concrete slab made of MPD3 and MPD4 girders post-tensioned by tendons. Based on the 2006 and 2007 diagnostic surveys, laboratory tests, normative regulations and recommendations and, last but not least, sensitivity analyses, an inverse design of selected design parameters was performed for required limit states. Various load levels, different alternatives of design parameters and different neural network structures were studied.
|
|
|
Life-time assessment of bridge based on nonlinear analysis and degradation modelling
Kucek, Martin ; Lehký, David (referee) ; Novák, Drahomír (advisor)
The aim of this work is to calculate the lifetime period of the bridge construction based on the nonlinear analysis and the degradation modelling. As a solution was the probabilistic nonlinear finite element analysis. The nonlinear numerical analysis is provided by the programme ATENA, developed by the Červenka Consulting. We have modeled one prefabricated bridge beam KA-73, which was overloaded by power effects until the structure was damaged. Observed limit states, limit state of the decompression, crack formation and limit state of the bearing were compared with characteristic material. It is possible to determine the maximum weighted capacity for the given construction based on discovered values.
|
|
|
Inverse reliability analysis of prestressed bridge
Lipowczan, Martin ; Novák, Drahomír (referee) ; Lehký, David (advisor)
The proposed diploma thesis deals with the application of methodology and tools of inverse analysis for design of selected structural parameters using a fully probabilistic analysis to determine the level of its reliability. The method based on artificial neural networks is used to approximate the inverse function. The inverse analysis was carried out in two ways that differs in the method of obtaining reliability indicators. The structure analyzed in this work was an existing bridge. The year of construction is estimated approximately between the years 1955 to 1960. The bridge is located close to the Uherský Ostroh. It is a one-piece concrete slab made of MPD3 and MPD4 girders post-tensioned by tendons. Based on the 2006 and 2007 diagnostic surveys, laboratory tests, normative regulations and recommendations and, last but not least, sensitivity analyses, an inverse design of selected design parameters was performed for required limit states. Various load levels, different alternatives of design parameters and different neural network structures were studied.
|
|
|
Life-time assessment of bridge based on nonlinear analysis and degradation modelling
Kucek, Martin ; Lehký, David (referee) ; Novák, Drahomír (advisor)
The aim of this work is to calculate the lifetime period of the bridge construction based on the nonlinear analysis and the degradation modelling. As a solution was the probabilistic nonlinear finite element analysis. The nonlinear numerical analysis is provided by the programme ATENA, developed by the Červenka Consulting. We have modeled one prefabricated bridge beam KA-73, which was overloaded by power effects until the structure was damaged. Observed limit states, limit state of the decompression, crack formation and limit state of the bearing were compared with characteristic material. It is possible to determine the maximum weighted capacity for the given construction based on discovered values.
|
guest ::
