Keywords: genetic algorithm, suspension bridge, reliability, performance, uncertainty, sensitivity analysis.

For their importance and their dimensions of the structure, the proper design of long suspension bridges is one of the greatest challenges for modern structural engineering. Nowadays, the performance based design theory drives the designers toward the solution of the design problem in which the most important performance indicators of the structure are considered [1,2]. In view of that, the performance of the structure must be evaluated with great accuracy both in static and in a dynamic way.

In traditional analysis the performance evaluation is based on a certain number of numerical simulations [3]. Every simulation is made to explore a particular performance indicator of the structure. The maximum vertical displacement or the maximum longitudinal slope, are explored using a few particular loads combinations to maximize or minimize the specific performance. In a complex structure, such as a long suspension bridge [4], to use traditional loads combinations may be insufficient to evaluate accurately the static and deformability parameters.

In this work, some static and deformability parameters are investigated using a genetic algorithm approach [5]. In particular, the following are investigated:

• the maximum vertical displacement;
• the maximum longitudinal slope;
• the maximum transversal slope;
• the maximum tension in the main cable; and
• the maximum tension in the tower legs.

For every parameter, a genetic algorithm analysis sets the worst load combination considering both vertical and longitudinal loads. Tens of thousands of different load combinations are considered for every genetic algorithm evaluation. In this way it is also possible to investigate the global deformability of the bridge, using the envelope diagrams from the analyses.

It is important to note the importance of the genetic algorithm approach in the analyses of the behavior of complex structures. This approach provides the analysts with the following possibilities:

• to perform stochastic exploration of the loads space;
• to define the worst load combination;
• to investigate the punctual value of a generic performance;
• to investigate the global behaviour of the structure through the definition of the envelope diagram of the performance;
• to handle uncertainties related to the definition of the loads.

Therefore, the genetic algorithm approach it is an important theoretical tool for the analysis and design of complex structures. The application developed in the paper provided the worst loads combination and the envelope diagram for each performance parameter investigated.

1

F. Bontempi, L. Catallo and L. Sgambi "Structural analysis and design of long span suspension bridges with regards to nonlinearities, uncertainties, interactions and sustainability", Proceedings of the IABMAS'04 Conference, Kyoto, Japan, 2004.

2

F. Bontempi, L. Catallo and L. Sgambi "Performance-based design and analysis of the Messina Strait Bridge", Proceedings of ASRANET 2004, Barcelona, Spain, 2004.

3

N.J. Gimsing "Cable Supported Bridges, concept and design", John Wiley & Sons, New York, 1983.

4

"Stretto di Messina S.p.A." Company, Preliminary Project, 1992, available on http://www.strettodimessina.it

5

L. Sgambi (In Italian) "I metodi dell'intelligenza artifi-ciale nell'analisi e nella progettazione di ponti sospesi", Ph.D. Thesis on Civil Engineering, University of Rome "La Sapienza", Advisor Prof. F. Bontempi. Available on-line http://padis.uniroma1.it/getfile.py?recid=267

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)