Participants
| Project leaders | Prof. Tom Lahmer, Prof. Dr. Guido Morgenthal |
| Project coworker | Dr. Tajammal Abbas |
| Project period | 01.07.2017 – 30.07.2020 |
| Project partner | DFG |
Goals
1. The essential wind excitation phenomena and the underlying stochastic parameters have to be determined for the considered structures. The limit states have to be defined that should form the basis of a probabilistic assessment of the structure depending on the dynamic response. This requites to develop strategies which allow an evaluation of the dynamic structural responses in consideration to the defined limit states.
2. Identify system parameters that will be the target variables of a structural optimization, as they have a significant influence on the aerodynamic behavior of the structure.
3. To adapt and use efficient mathematical algorithms for robust structure optimization in order to optimize the structure with respect to the defined limit states. In the context of a computational framework to be implemented, the interface between individual components have to be developed which produces accurate and efficient results. Optimization results and variations are then to be visualized.
4. The optimized designs are to be tested in the wind tunnel for their actual suitability on a model scale. The computational challenges of numerical simulation and stochastic optimization are very high. Therefore methods can be utilized which can reduce the computational effort by means of sensitivity analyses, metamodeling and phenomenon-based modeling.
Description
The aeroelastic behaviour of long-span cable-supported bridges is essential to be studied as part of the design and analysis. These structures could develop significant vibrations when exposed to atmospheric wind flow. The trends for increase in the flexibility and reduction of mass of structures make such problems more prominent and the analysis more challenging; therefore, it is necessary to keep a balance between safety and performance. This also demands the use of accurate and robust prediction models. Typically, experimental methods are used for the practical design of structures because of their accuracy; however, these methods are time consuming and expensive. Numerical methods have gained much attention and development in the last decades and are used as alternative beside the analytical and experimental methods. However, numerical methods still have limitations with the complexity and nature of three-dimensional flow around structures. Nevertheless these methods can provide the underlying physical processes to better understand the complex aeroelastic phenomenon.
The fundamental aim of this project is to develop a framework which makes it possible to evaluate the forces of structures from complex dynamic wind-induced phenomena by means of numerical simulations with defined limit states and to optimize the structure on the basis of the results such as probability of occurrence which must remains below a certain threshold. The underlying simulation methodology limits the application to line-like structural elements, such as bridges, skyscrapers, towers, masts or chimneys. The characteristic of the natural wind field results in a problem with stochastically defined input parameters.
The simulation methods and optimization strategies are to be selected, tested and adapted. Additionally, interfaces and evaluation algorithms for efficient computation must be developed and tested. It is hoped that such methods can be feasible for resolving complex optimization problems using this framework in the context of practical application to structures.