Improvement of High Lateral Resolution Surface Wave Method with Frequency-Wavenumber Filter

• Main Authors: zheng, ting, 鄭挺
• Other Authors: 林志平
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/59568116464113246469

碩士 === 國立交通大學 === 土木工程系所 === 104 === The surface wave seismic method has gained popularity in engineering practice for determining shear wave (S-wave) velocity depth profiles after few decades of development. However, there are tradeoffs caused by various survey line parameters (such as near offset, receiver spacing, and offset range) in traditional surface wave test. Currently, better results can be obtained by applying high resolution surface wave method (data combination of multiple source), nonetheless in some lateral heterogeneous layers, using phase-seaming method to combine data of different sources may cause dispersion curve jumping. This study hence aimed at two objectives: (1) Exploring possible combination of survey line parameters and lateral heterogeneous underground conditions which would result in dispersion curve jumping; (2) Introducing a corresponding feasible solution for bandwidth optimization on dispersion curve. This study adopted a fourth-order velocity-stress finite difference method to simulate data based surface wave testing pseudo-section survey lines in different earth models with lateral heterogeneity. According to the simulation result, this high resolution surface wave method is mainly applicable to single-mode dominant surface wave signals. Therefore, in layer conditions which may generate more than single mode, the proposed concept would be erratic due to dispersion curve jumping after combination. Another approach proposed is to analyze by f-k (frequency-wavenumber) filter to filter out different modes and proceed with aforementioned data seaming and scanning procedure. By observing the results of filtered data, the feasibility of f-k filter is determined and high resolution signal processing of surface wave method optimization is then established in this study. This study proposed an optimization concept by applying a variety of frequency domain bandwidth surface signals, in order to achieve the best dispersion curve bandwidth based on high resolution surface wave method, comprising of f-k filter and data combination technique. Field results demonstrated the high feasibility of this application, in which the data combination of various effective energy bandwidth signals not only increased the bandwidth of dispersion curves, but also greatly enhanced the resolution of dispersion curve profile. This finding would boost the efficiency and data quality of field surveys and can be included in standard operation procedures of field surveys in near future.