Practical Engineering Design Tool for Vibration Sensitive Laboratory Building Structures

Abstract

The prediction of structural vibrations in vibration sensitive laboratory structures is a complex problem involving computationally heavy 3D models with long computational times. However, in the early design phases of a project long computational times are undesirable. Therefore the problem statement of this Master thesis concerns the development of a practical engineering tool which can be used in the early design phase of a vibration sensitive laboratory structure. The resulting tool is called EDDABuSgs (Early Design Dynamic Analysis of Building Structures by Gerwin Schut) and is the final product of the thesis research. The main focus of EDDABuSgs is on structural vibrations induced by heavy traffic at geological locations with soft soils (e.g. Amsterdam, the Netherlands). The thesis research concerns the four interrelating parts: source of vibrations (truck), transmission of vibrations (soil), soil-structure interaction and the structure (receiver of vibrations). The software FEMIX is used for computing the 3D soil response (source and transmission). The soil response is then used as input for EDDABuSGS, which computes the 2D structural vibrations by the aid of a Python script. The global structural response is computed from a rigid 3DoF system which is supported elastically by the soil and a pile foundation. The local structural response (ground floor) is computed from a flexible frame, composed of the analytical solutions of Euler-Bernoulli beam elements. The flexible frame is excited by the global 3DoF structural response by means of the boundary- and interface conditions. The results of FEMIX and EDDABuSgs show good agreement with the used verification projects. Several iterations have been made using EDDABuSgs to see how several parameters change the 2D structural response. The results of these iterations are well in line with general known theory about structural dynamics. From this Master thesis research one can conclude that soft soils excited by heavy traffic have a responsive frequency spectrum generally in-between 3 and 15 Hz. Therefore it generally holds that the eigenfrequencies of the global building response should be made relatively low (< 3 Hz), while the eigenfrequencies of the local structural elements (e.g. floors) should be made relatively high (> 12 Hz). Additionally, the resistance against vibrations (impedances) of the structural elements should to be as large as possible, which might sometimes contradict the preferred shift of the eigenfrequencies.