寒区隧道冻结模型试验与温度场特性分析

发布时间：2018-03-26 23:01

本文选题：人工冻土　切入点：模型试验　出处：《安徽理工大学》2017年硕士论文

【摘要】：以青藏高原为代表的高海拔寒区隧道在使用过程中出现的冻害现象导致隧道的使用率大大降低甚至报废。由于冻害的影响,国家投入巨资无法收回,隧道通行人民的生命安全也无法保障。本文以寒区隧道冻害为背景,开展了人工冻土模型试验,并针对试验数据的有限性的特点,利用基于结构风险最小化原理和小样本学习方法——支持向量机算法,建立了人工冻土温度场发展的支持向量机模型。1、通过寒区隧道模型试验,得到了寒区隧道模型温度变化规律:寒区隧道模型温度变化分为两个阶段:降温段和升温段,本文暂时不考虑升温段的变化规律。降温段变化规律大致为:随着冻结时间的增加,围岩各测点的温度不断降低,不同断面、不同测点的温度降低速率有些差异,整体来看可以分为断面3和其他断面两组。2、根据牛顿热交换定律:物体从一种介质流到另一种介质的热量和两种介质间的温差成正比,初始冻结时,隧道围岩为室内温度与冻结管的温差最大,故在此时降温速度最快,随着冻结时间的增长,两者间的温差减小,故在此时降温速度减慢;到冻结后期时,两者的温差很小,故隧道围岩温度不再变化,趋于一个稳定值。3、位置3的降温大致呈现一种对数函数减小的规律。其原因是位置3的温度传感器布设较多,布设比较均匀,反应隧道围岩温度变化比较准确;其他位置的的降温规律大致呈现线性减小趋势。4、通过对SVM法的介绍和推导建立了人工冻土模型试验温度场SVM模型。5、把不同核函数应用到所建模型上,通过比较选出最适合的核函数,然后对温度场温度进行预测,通过与实测数据的对比,发现预测值与实测值相差不大,验证了所建模型的合理性与正确性。
[Abstract]:The freezing damage of tunnels in high altitude cold regions, represented by the Qinghai-Tibet Plateau, has led to a great reduction in the utilization rate of tunnels and even their scrapping. Because of the impact of the freezing injury, the state has invested a huge amount of money that cannot be recovered. The safety of tunnel people's lives can not be guaranteed. In this paper, the artificial frozen soil model test is carried out under the background of tunnel freezing injury in cold region, and the characteristics of the limited test data are pointed out. Based on the structural risk minimization principle and the support vector machine (SVM) algorithm, a support vector machine (SVM) model for the development of temperature field in artificial frozen soil is established, which is based on the principle of structural risk minimization and small sample learning method. The temperature variation of tunnel model in cold region is divided into two stages: the cooling section and the heating stage. In this paper, the variation law of the temperature rise section is not considered for the time being. The variation law of the cooling section is: with the increase of freezing time, the temperature of each measuring point of surrounding rock decreases continuously, and there are some differences in the temperature decreasing rate of different measuring points with different sections. As a whole, it can be divided into two groups: section 3 and other sections. 2, according to Newton's law of heat exchange: the amount of heat flowing from one medium to another is proportional to the temperature difference between the two media, initially frozen. The temperature difference between the surrounding rock of the tunnel and the freezing pipe is the largest, so the cooling speed is the fastest at this time, and with the increase of freezing time, the temperature difference between the two decreases, so the cooling speed slows down at this time, and the temperature difference between the two is very small at the later stage of freezing. Therefore, the temperature of surrounding rock of tunnel is no longer changing, and tends to be a stable value. 3. The cooling of position 3 shows a law of decreasing logarithmic function. The reason is that the temperature sensors of position 3 are arranged more evenly. The temperature change of reaction tunnel surrounding rock is more accurate; The law of cooling in other places shows a linear decreasing trend. Through the introduction and derivation of SVM method, the temperature field SVM model of artificial frozen soil model is established, and the different kernel functions are applied to the established model. By comparing and selecting the most suitable kernel function, the temperature field temperature is predicted. By comparing with the measured data, it is found that the predicted value is not different from the measured value, which verifies the rationality and correctness of the model.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U456

【参考文献】