Study On Dynamic Characteristics Of Slide-zone Soil And Stability Of Loess Landslides Under Train Vibration Load

In recent years, with the development of railway modernization, raising train-running speed and increasing train-hauling load will necessarily increase train vibration force, the vibratory effect of train-running on stability of loess landslides has become a hot issue among Geotechnical engineering insiders. In the thesis, taking the loess landslides and slide-zone soil along the railway as researched subject and combining with geo-technical investigation, laboratory tests, mechanical analysis, theoretical study, numerical calculation, the dynamic characteristics of slide-zone soil and stability of loess landslides under high-speed train vibration load is studied deeply. The main research results of dissertation can be summarized as follows:(1) The cause of formation and characteristics of loess landslides under train vibration load is studied deeply. Through the analysis, train vibration load is thought to be a chief factor which induce the landslides; The characteristics of slide-zone soil is Summed up, and based on the theoretical foundation for laboratory tests of slide-zone soil.(2) By imposing the different cell pressure and dynamic load on slide-zone soil specimens which collected along the typical railway landslides in dynamic tri-axial test, the rules about dynamic strength and dynamic deformation of the slide-zone soil under simulating the train vibration load are studied. The results show that the dynamic strength and cumulate residual strain of the slide-zone soil change with the change of dynamic time and consolidation stress under certain dynamic stress.(3) Based on this test data about slip zone soil dynamic tri-axial test, the BP neural network forecasting model of slip zone soil vibratory subsidence is established, and the results are compared with the multivariate linear regression model. The results show that the assessment of BP neural network forecasting model is more accurate than multivariate linear regression model.(4) The stability of loess landslides under high-speed train vibration load is analyzed by using FLAC. The results indicate that disturbed degree and stability coefficient of the loess landslides change with different train-speed. The results are helpful for the design and strengthening for high-speed railway loess landslides.(5) Through the study about mechanism of loess landslides, the mechanism of loess landslides under minor amplitude and long-term vibration load is proposed. And analysis based on the principle of mechanics and entropy is carried out to prove the proposed mechanisms.