【作者】 刘钢；

【导师】 罗强；

【作者基本信息】 西南交通大学 ， 道路与铁道工程， 2013， 博士

【摘要】 高速铁路对轨道结构提出了高平顺性、高稳定性的要求。严格控制路基工后沉降,为轨道结构提供一个坚实、稳固的基础,是实现轨道结构长久保持高平顺性和高稳定性的先决条件,也是高速铁路保持其长期服役性能的关键问题之一。作为路基结构上部的基床部分,直接承受列车长期动荷载作用和气候环境变化的影响。在高周循环荷载作用下基床结构的累积塑性变形是线路结构长期变形的重要组成部分,以控制长期累积塑性变形为出发点,建立基于力学分析和满足功能要求的基床结构设计方法不仅是对现有基床结构设计理论的完善与发展,对已建高速铁路基床结构在未来数十年内是否能保持其长期良好服役性能的评价也具有重要意义。在吸取、总结和分析已有研究成果基础上,以构建基于长期累积变形演化状态控制的基床结构设计方法为核心,开展了列车荷载经轨道结构传递至路基面的动应力作用模式、土工填料在循环荷载作用下的长期累积变形演化状态特征及关键控制参数、基床结构的室内大型动态模型试验、路基压实质量检测中的小型平板载荷试验技术等方面研究。主要工作和结论如下：1、基床结构承受的列车荷载作用特性分析无砟轨道具有刚度大、稳定性好、传递扩散荷载能力强等特点,固定轴距的两个轴载之间存在明显的叠加效应,实测路基面动应力时程曲线表明,无砟轨道路基面动应力一次加卸载过程由一个转向架的两个轴载共同作用完成,而有砟轨道路面承受列车荷载一次加卸载作用由单个轴载完成。基于实测的无砟轨道路基面动应力时程曲线,建立了动应力随时间变化规律与列车轴载移动的空间位置关系,确定了列车荷载经轨道结构传递扩散至路基面纵向影响范围的计算方法,得到的纵向影响距离在9～10m之间,与Winkler弹簧地基上叠合梁的理论计算结果吻合。在此基础上,提出了无砟轨道路基面纵向梯形、横向均匀荷载作用模式。针对目前有砟轨道路基面荷载作用模式采用2.8m纵向影响距离未能考虑轨枕间距、道床厚度等因素的不足,以单轮载力由5根轨枕分担为基础,假定列车荷载在道床中沿线路纵向按“折线型”扩散的模式下,建立了有砟轨道路基面动应力纵向影响范围与轨枕间距、道床厚度之间的关系式,进一步完善了有砟轨道路基面动应力影响范围的确定方法。通过建立列车轴载与传递扩散至路基面动应力的力学平衡关系,分析综合动力影响系数的取值及考虑有砟轨道路基面动应力横向不均匀影响修正的基础上,得到路基面动应力计算方法及用于基床结构设计中承载力和耐久性检算的荷载水平。通过对无砟轨道路基面实测动应力时程曲线进行幅频特性分析表明,无砟轨道对车辆荷载的良好扩散作用已致使转向架轴距对应的频响特征不明显,大幅降低了路基承受的动荷载作用频率,主要作用频率分别是以车长为基本扰动波长计算频率值的1、2、3、4倍。2、循环荷载作用下土工填料长期累积变形演化状态特征及填土单元模型循环加载试验研究基于循环荷载作用下土工填料变形规律分析及有关累积塑性变形发展状态的已有研究成果,以累积塑性变形演化呈现出收敛与否、收敛发散快慢的现象特征为出发点,提出了循环荷载作用下土工填料累积塑性变形的演化呈现快速稳定、长期稳定、长期破坏和快速破坏的四个状态类别,建立了基于累积塑性变形速率发展趋势,以负幂函数f(N)=CN-p表达的数学判别准则(p值判别法),当p≥2时为快速稳定状态、1<p<2时为长期稳定状态、0<p≤1时为长期破坏状态、p≤0为快速破坏状态。构筑了压实系数分别为0.9、0.95和1.0条件下的级配碎石填料模型,进行了直径为30cm的刚性承载板循环加载试验及地基系数K30测试,基于“p值判别法”得到了级配碎石填料对应快速稳定、长期稳定、长期破坏和快速破坏四个状态的三个循环应变阈值。试验表明：压实系数0.9、0.95和1.0状态所对应的级配碎石填料K30值分别为137MPa/m、214MPa/m、380MPa/m;由快速稳定过渡为长期稳定状态的循环应变阈值εt1≈140～202με、均值εl1约为164με,由长期稳定转化为长期破坏状态的εt2≈443～650με、εt3约为526με,由长期破坏发展至快速破坏状态的εt3≈1339～2172με、εt3约为1734με；建立了循环应变阈值和动承载力阈值与K30值之间的经验关系式。3、基于长期累积变形演化状态控制的基床结构设计方法及其关键技术指标基于建立的基床结构设计荷载作用模式及获得的基床填料长期累积变形演化状态控制参数,以满足不同轨道结构对基床变形状态控制要求为目标,采用循环应变为核心控制指标、K。值为核心设计指标,运用力学分析原理,构建了基于长期累积变形演化状态控制的高速铁路基床结构设计方法及设计流程,提出了与有砟轨道和无砟轨道基床结构循环应变水平状态相适应的填料参数取值原则。运用基床结构长期累积变形演化状态控制设计方法,探讨了不同基床表层厚度条件下,无砟和有砟轨道结构基床底层的最小K30。控制值。对于无砟轨道结构,设计K30值为190MPa/m的基床表层厚度在0.2m-1.2m时,基床底层最小K30控制值约为108MPa/m～102MPa/m;相同基床表层条件下的有砟轨道结构,基床底层的最小K30控制值约为100MPa/m～68MPa/m。由此可见,无砟轨道结构需要较高基床底层填筑压实标准以满足长期累积变形快速稳定的控制要求,而有砟轨道结构的基床底层K30控制标准受基床表层厚度的影响明显。4、基床结构室内大型动态模型试验分析在室内构筑大型填土模型,进行了基床结构的大型动态模型试验。模型填土平面选取3m×3m的单元结构,厚度为2.2m,砖墙和砂袋堆砌组成的边界约束,模拟半无限空间体中局部结构在循环荷载作用下的力学响应。对模型结构施加了荷载水平为20kPa、40kPa、60kPa、80kPa、100kPa、120kPa,共计260万次的循环加载试验。试验过程中,测试了模型结构不同深度出的弹性变形、累积塑性变形、动应力、以及距顶面40cm的侧向弹性和累积塑性变形。当荷载水平小于60kPa时,最大循环应变为166με,沿深度范围内的循环应变均小于模型填土压实状态所对应的循环应变阈值ε1,处于快速稳定状态；当荷载水平为100kPa和120kPa时,模型表层约50cm厚度范围内的循环应变超过了应变阈值ε,,推断该结构层范围内累积塑性变形演化处于长期稳定状态,实测的累积塑性变形的演化趋势很好的映证了这一结论。试验结果验证了基于长期累积变形演化状态控制的基床结构设计理念的合理性及以循环应变为关键控制技术指标的可行性。5、铁路路基压实质量检测中的地基系数Ko、变形模量E2试验技术分析基于室内填土模型试验,对比分析了变形模量、地基系数K30两种小型平板载荷试验的特点,重点讨论了“沉降稳定控制法”和“等时间间隔控制法”两种加载控制方式对试验结果和试验效率的影响。结果表明：K30较Ev2更能直接反映路基的压密程度,每级荷载的保持时间对试验结果有显著影响,满足1%沉降稳定控制标准需较长的试验历时；K30试验的沉降稳定控制标准由1%改为2%、或加载时间间隔为6min,可缩短30%～50%试验时间,K30值误差在10%以内；对两种加载控制方式试验结果分析基础上,提出了基于荷载与沉降稳定时间呈正相关性的“荷载-时间控制法”试验加载方式,即每级荷载保持时间Δt1(min)与加载级数i(i≥2)相等且Δt1＝2min.该方法可提高试验效率近50%、且具有可操作性强、误差小等优点,用于铁路路基砾石类填料压实质量的K30。检测具有良好适应性。更多还原

【Abstract】 The high-speed railway requires the track structure with high smoothness and stablity. Controlling the post-construction settlement of subgrade is the necessary prerequiaite that keeps high smoothness and stablity of track, which is also the key factor of the high-speed railway maintaining good long-term service performace. The cumulative plastic deformation under numerous cyclic loading is an important component of the post-construction settlement for the upper structure of subgrade. Starting with the controlling of the cumulative plastic deformation, the design method of subgrade structure was found which not only is the development and improvement of the existing subgrade structure design theories, it would be significant for assessing the service perfprmance of high-speed railway subgrade which has constructed and put into sevice in it’s design working life, about100years.Summing up the results of research at home and aboard, the paper circled around to construct the subgrade structure design method based on the long-term cumulative defornation evolution state controlling, the characteristics of subgrade bearing train load and the controlling parameters of subgrade filling and the indoor large scale dynamic model testing of subgrade and the test technique of the coefficent of subgrade reaction were studied. The main work and conclusions are as follows:1. Analysis on the characteristics of the subgrade bearing train loadAs ballastless track structure has bigger stiffness, better Stability and better capacity for diffusing load, the two axle loads between the bogie wheelbase have obvious superimposed effect. The dynamic stress time history curve which is measured on the subgrade surface shows that one loading and unloading process of the dynamic stress in ballastless track subgrade is completed by two axle loads of the same bogie. While for ballasted track, the process is completed by single axle load.The relation between the changing regularity of dynamic stress with time and the movement of the train axle load is established which based on the the dynamic stress time history curve which is measured on the subgrade surface of ballastless track. And the calculation method of longitudinal impact range of train load which diffuses to the subgrade surface through the track structure is determined b. What’s more, by using that calculation method, the calculated longitudinal impact range is between9m and10m, which is consistent with the theoretical calculation result of Winkler spring foundation composite beam theory. Based on these theories and methods, the load model of ballastless track and the load distribution can be simplified to be trapezoidal distribution in the longitudinal of the lines and to be uniform distribution in the transverse direction, is proposed. In the existing design, the load model of ballasted track supposes the longitudinal impact range of train load is2.8m. Here, Sleeper spacing, thickness of ballast and other factors are not concerned. Aimed at the above shortcoming and based on the assumption that single wheel load is shared by the5sleeper and diffusion model of the train load is in broken line form, the relationship formula among the longitudinal impact range of dynamic stress on the subgrade surface in ballasted track, the distance of sleepers and ballast thickness is established. So, the ways to determine the impact range of dynamic stress on the subgrade surface in ballasted track is furture perfected. By establishing the equilibrium relationship between the train axle load and dynamic stress on the subgrade surface, analysing the comprehensive coefficient values of dynamic effect and considering the correction of transversely uneven effect of dynamic stress on the subgrade surface in ballasted track, the calculation method of dynamic stress on the subgrade surface and the load level for checking load-bearing capacity and durability in the roadbed design are got.The amplitude-frequency characteristic analysis for the dynamic stress time history curve,which is measured on the subgrade surface of ballastless track, shows that the good capacity for diffusing train load in ballastless track makes the frequency response characteristics, which is corresponding to the bogie wheel base, not obvious. The frequency of subgrade under dynamic load is sharply reduced. And the main frequencies are1,2,3,4times of the frequency value which is calculated by using the train length as the basic disturbance wavelength.2. The status features of the long-term cumulative deformation of the geotechnical filling under the cyclic loading and the study of the cyclic loading research of the filled earth unitThe analysis of the deformation law of the geotechnical filling under cyclic loading and the study of the development status of the cumulate plastic deformation has obtained results. Based on the existing research results, with the start point that whether the cumulate plastic deformation converge or not and whether the speed of convergence is fast or slow, the paper put forward that there are four status of the cumulate plastic deformation development. They are fast settling, long-term settling, long-term breaking and fast breaking. What’s more, the paper establish the criterion based on the development of the cumulate plastic deformation speed, which is expressed by the negative power function:f(N)=CN-P. WhenP≥2, it’s called the fast settling status; when1<P<2, it’s called the long-term settling status; whenO<P≤1, it’s called the long-term settling; whenP≤0, it’s called the fast broken status.It constructs the model of the graded broken stone with the different compacting factor of0.9,0.95,1.0and conducts the cyclic loading experiment which can set out the ground coefficient K30with the rigidity loading board. Then, it obtains the three threshold value of the four status of the graded broken stone which called fast settling, long-term settling, fast broken and long-term broken based on the "p-convergence". The experiment indicates that the K30value of the graded broken stone of the different compacting factor of0.9,0.95,1.0is137MPa/m,214MPa/m,380MPa/m respectively. The transition threshold value ε11from the fast settling status to the long-term settling is140-202μs, the mean value ε11is64με. The transition threshold value ε12from the long-term settling status to the long-term broken is443-650με, the mean value ε12is526με. The transition threshold value ε13from the long-term settling status to the long-term broken is1339-2172με, the mean value ε13is1734με. Moreover, it establishes relationship between the threshold value of the cyclic strain, dynamic bearing capacity and the K30value.3. The design method for subgrade of the high-speed railway based on the deformationstate controlling and the key technical indexsBased on established load pattern of bed structure design and bed filing accumulated deformation evolution status control parameters, In order to satisfy different track structures on bed deformation state control requirements, Using cyclic strain as the core control indicators、K30values as the core design index, Using the principle of mechanics analysis, Constructed based on the long-term cumulative deformation evolution state control of high speed railway bed structure design method and design process, Put forward ballast track and ballastless track bedding structure of cyclic strain corresponds to the level state of filling parameter selection principle.Using long-term cumulative deformation evolution state control bed structure design method, Discussed under the condition of different thickness of bed surface, ballast track and ballastless track structure at the bottom of bed the minimum control K30values. For ballastless track structure, Design K30values for190MPa/m the surface layer of bed thickness in0.2m-1.2m, at the bottom of bed the minimum control values of K30about108MPa/m-102MPa/m; The same under the condition of bed surface ballast track structure, at the bottom of bed the minimum control values of K30is about 100MPa/m～68MPa/m. Thus it can be seen ballastless track structure need the higher at the bottom of bed fill compaction standards to meet the long-term cumulative deformation fast and stable control requirements, while ballast track structure at the bottom of bed the control standards of K30significantly influenced by the thickness of the surface layer of bed.4. The large scale dynamic model testing of subgrade in the laboratoryThe large-scale made ground mode is set up indoor, the large dynamic test is taken. This model, which is consisted of made ground and boundary constraint,could simulate the mechanical response simulation of local structure in elastic half space under cyclic loading. And the made ground has a surface of3m×3m and a thickness of2.2m,the boundary constraint is provided by brick walls and sand bag stack. The model structure is applied by a series of load level of20kPa,40kPa,60kPa,80kPa,100kPa,120kPa, giving a total of2600000times of cyclic loading. During the test, the elastic deformation,cumulative plastic deformation, dynamic stress of different depth of the model structure, as well as the lateral elasticity and the cumulative plastic deformation from the top surface of40cm will be tested.When the load level is less than60kPa, the maximum cyclic strain is166με, the cyclic strain along the depth is less than the cyclic strain threshold εt1which is corresponding to the compaction state of the made ground, being in a fast steady state; When the load level is100kPa and120kPa, the cyclic strain of the range from the model surface about50cm thickness is bigger than the strain threshold, infering that the evolution of the cumulative plastic deformation within this structure layer is in a long steady state.This conclusion is verified well by the evolution trend of real measured deformation cumulative plastic. The test results show the rationality of the subgrade design concept based on the control of the long-term evolution deformation and the feasibility of the cyclic strain as the key control index.5. The test technique of the coefficient of subgrade reaction in the compaction quality detection of subgradeBased on the model test of laboratory fill, analyze and compare the characteristics of the two mini-plate loading test through the indoor testing of deformation modulus and coefficient of foundation K30, and particularly focusing on the influence and efficiency of the loading mode upon the testing results. Results show that, the ground coefficient can better reflect the compaction quality of subgrade than the strain modulus. The time interval of loading has significant influence on the testing results. The testing will take a long time to satisfy the steady criterion of deformation1%. And the testing time will reduce about30%-50%if the deformation steady criterion changed from1%to2%, or the time interval of loading was about6min, and the K30error will within10%. On the basis of analysis experimental results of the loading mode, the load-time controlling loading method was proposed, considering load is in positive correlation with the settlement stability time, namely the time interval of loading, Δt1=2min, is equal to load grade. The method improves test efficiency nearly50%, and has advantage of easy operation and small errors, which is especially suitable for the compaction quality detection of railway subgrade being made of gravel soil.更多还原