Transfer Function Prediction Methods For Environmental Influence Of Metro Train-induced Vibrations

With the development of urban rail transit network, the influence on the nearby environment caused by metro train vibration becomes more and more remarkable and extensive, including residents, buildings, precision instruments and so on. Some vibration sensitive objects even become the controlling factors in the entire metro-line construction progress. Therefore, it is very important to predict accurately and scientifically the vibration effect on different sensitive objects in different construction stages.To that end, and supported by the National Natural Science Foundation of China "Study on the Prediction Model for Metro-induced Ambient Vibrations Based on In-situ Drill Impact"(51278043) and "Study on Prediction Model and Mitigation Laws of Environmental Vibrations Induced by Variable-speed Metro Trains in Curved Tunnels"(51378001), new transfer function prediction methods to predict environmental influence of metro train-induced vibrations are proposed. Moreover, to meet the requirements of different prediction accuracy required by metro construction in feasibility study stage, preliminary design stage and construction design stage, three new implementation methods are proposed accordingly:the analytical transfer function fast prediction method, the deep-hole-excitation measured transfer function prediction method and the in-tunnel-excitation measured transfer function prediction method, and each implementation methods’prediction theories and processes are proposed. Meanwhile, a systematic study has been conducted for the key issues of the new established transfer function prediction methods. A large number of research works of theoretical analysis, laboratory tests and numerical calculations are implemented to acquire a series of valuable conclusions.During the four years of research work, the following five major innovative research works were carried out:(1) the classification of metro-train-induced environmental vibration prediction methods has been studied. From the perspective of different construction stages, all kinds of metro-train-induced environmental vibration prediction methods are re-sorted according to the level of prediction accuracy, and the existing prediction methods are classified as:preliminary prediction, confirmed prediction and accurate prediction. (2) The transfer function prediction system for environmental influence of metro train-induced vibrations are established, including three different implementations:the analytical transfer function fast prediction method, the deep-hole-excitation measured transfer function prediction method and the in-tunnel-excitation measured transfer function prediction method. These prediction methods can consider phase information comprehensively. The prediction principles of different methods are studied, and the implementation processes of their theories are derived. The transfer function prediction methods are combined with the standard (HJ453-2008) recommended chain attenuation empirical prediction method and traditional confirmed prediction methods to construct a comprehensive dynamic prediction system for all metro construction stages. (3) The deep-hole-excitation device for in-situ deep-hole-excitation tests is developed. This device can meet the technical requirements of deep-hole-excitation test to implement the excitation test for any metro design depth, and achieve high-precision sampling of vibration source input force and ground vibration response signals. (4) In-situ vibration excitation tests have been carried out several times. The applicability and prediction accuracy of transfer function prediction methods are verified. (5) Based on the three-dimensional dynamic finite element simulation model with/without tunnel structure, the influence of tunnel structures on vibration transfer function is studied considering four different buried depth. A frequency domain correction function of tunnel structure is defined and the calculation method is proposed and the frequency domain correction functions of tunnel structure required by the deep-hole-excitation measured transfer function prediction method are solved.