【作者】 李向阳；

【导师】 崔新壮；

【作者基本信息】 山东大学 ， 土木水利（专业学位）， 2023， 硕士

【摘要】 我国重载铁路属于货运专线铁路的特殊类型、顶级种类,主要运输大型货物,是我国大宗货物运输的主要渠道之一。与普速客货共线铁路相比,重载铁路运输具有轴重大、轮组多的特性,使路基工作区范围更广,在路基内部应力路径和应力主轴旋转现象更为复杂,土体细观结构演化更加剧烈,导致路基塑性变形累积速度与服役性能衰变更加严重,从而威胁列车安全运营。因此,本文依托国家重大科研仪器研制项目——重载交通荷载作用下地基累积变形原位模拟系统(52027813),建立了重载铁路路基动力响应计算模型,并使用大秦铁路现场路基土样开展了一系列动态三轴序列加载试验、动态三轴长期循环加载试验,建立了交通荷载作用下空心扭剪离散元数值模型,深入研究了重载列车荷载作用下铁路路基三维动力响应、宏观动力特性和细观演化机理,建立了土体动回弹模量和累积变形预测模型,主要研究成果如下:(1)基于扩展Biot非饱和土理论,建立了轨道-轨枕-道砟-非饱和路基耦合动力模型,获得了重载列车荷载作用下路基三维应力分布和土单元应力路径,并利用现场测试验证了模型的正确性。研究表明,土单元竖向位移与饱和度、列车速度和轴重呈正相关,与土单元深度呈负相关;其各方向应力与列车轴重呈正相关,与饱和度和土单元深度呈负相关;水平正应力σx和σy与列车速度成反比,但列车速度对竖向正应力σz和剪应力τxz无明显影响。值得注意的是随着深度增大,各方向应力曲线波形逐渐由“多峰”转变为“单峰”,应力主轴旋转的次数也随之减少。(2)基于理论计算,利用大秦铁路平谷段路基土样开展了一系列重载铁路应力水平下的动三轴序列加载试验和长期循环加载试验。试验发现土体动回弹模量与动偏应力幅值、含水率呈负相关,与围压、压实度呈正相关;试样的塑性变形随加载次数的增加而不断累积,且与动应力幅值、含水率呈正相关,与围压、压实度呈负相关。建立了动回弹模量及累积变形预测模型,并提出了基于遗传算法优化BP神经网络的动回弹模量预测方法。(3)基于DEM-FDM耦合方法,建立了完全柔性边界下的土体动态三维复杂应力路径加载数值模型,并通过空心扭剪单元体室内试验,标定了数值模型细观参数,对不同应力水平下的空心扭剪试验进行模拟。结果表明,空心圆柱单元体数值模型的宏观累积变形与加载应力、动剪应力呈正相关,与围压呈负相关;孔隙率、平均配位数、水平向接触数和法向接触力变化程度与轴重、动剪应力成正比,与围压呈反比。更多还原

【Abstract】 China’s heavy-haul railway is a special and top-level type of freight dedicated railway,mainly transporting large goods,and is also one of the main channels for bulk cargo transportation in China.Compared to the common high-speed passenger and freight railway,heavy haul railway transportation has the characteristics of significant axles and multiple wheel sets,which widens the scope of subgrade work area,complicates the stress path and rotation of the stress principal axis within the subgrade,and makes the evolution of soil microstructure more intense,leading to more serious deterioration of the cumulative speed of subgrade plastic deformation and service performance,and threatening the safe operation of trains.Therefore,relying on Special-funded Program on National Key Scientific Instruments and Equipment Development——In Situ Simulation System of Foundation Cumulative Deformation under Heavy Traffic Load(52027813),this article has carried out a series of studies with the heavyhaul railway as the research object.A calculation model for the dynamic response of the subgrade of the heavy-haul railway was established,and a series of dynamic triaxial sequential loading tests and dynamic triaxial long-term cyclic loading tests were conducted basing on soil samples of Datong-Qinhuangdao railway subgrade.Subsequently,a discrete element numerical model of hollow torsional shear under traffic loads was established,the three-dimensional dynamic response,macroscopic dynamic characteristics,and microscopic evolution mechanism of railway subgrade under heavy load train loads were deeply studied,and a prediction model for dynamic modulus of resilience and cumulative deformation of soil mass was established.The main research results are as follows:(1)Based on the extending Biot’s unsaturated soil theory,a coupled dynamic model of track sleeper ballast unsaturated subgrade was established,and the three-dimensional stress distribution and soil element stress path of the subgrade under heavy load train loads were obtained.Then the correctness of this model was verified by field tests.The research shows that the vertical displacement of soil element is positively correlated with saturation,train speed,and axle load,while negatively correlated with soil element depth;The stress in all directions are positively correlated with the axle load of the train,and negatively correlated with saturation and soil unit depth;Horizontal normal stress σx and σy is inversely proportional to the train speed,but the train speed is related to the vertical normal stress σz and shear stress τxz had no significant effect.It is worth noting that as the depth increases,the stress curve waveform in each direction gradually changes from "multi peak" to "single peak",and the number of rotations of the stress principal axis also decreases.(2)Based on theoretical calculations,a series of dynamic triaxial sequential loading tests and long-term cyclic loading tests under heavy haul stress levels were conducted using soil samples from the Pinggu section of the Datong-Qinhuangdao Railway subgrade.The test found that the dynamic resilient modulus of soil is negatively correlated with the amplitude of dynamic deviator stress and water content,but positively correlated with confining pressure and compactness;The plastic deformation of the sample continuously accumulates with the increase of loading times and is positively correlated with the dynamic stress amplitude and water content,while negatively correlated with the confining pressure and compactness.A prediction model for dynamic resilient modulus and cumulative plastic deformation is established,and a prediction method for dynamic resilience modulus based on genetic algorithm optimized BP neural network is proposed.(3)Based on the DEM-FDM coupling method,a dynamic three-dimensional complex stress path loading numerical model of soil mass with fully flexible boundaries was established.Through indoor tests of hollow torsional shear units,the mesoscopic parameters of the numerical model were calibrated,and hollow torsional shear tests at different stress levels were simulated.The results show that the macroscopic cumulative deformation of the numerical model for hollow cylindrical elements is positively correlated with the loading stress and dynamic shear stress,while negatively correlated with the confining pressure;The variation of porosity,average coordination number,horizontal contact number,and normal contact force is directly proportional to the axial load and dynamic shear stress,and inversely proportional to the confining pressure.更多还原