基于热示踪的饱和层状介质水热参数反演

Estimation of hydraulic and thermal parameters in saturated layered porous media based on heat tracing method

多孔介质的水热运移参数是影响介质中水热运动过程的主要因素。本文基于热示踪方法,开展了稳定流场条件下饱和层状石英砂的热示踪实验,分别采用热电偶与热成像两种方法测定了砂箱内部和表面的温度,同时结合HYDRUS模型的反问题算法对层状石英砂的水热运移参数进行了反演。研究结果表明:当介质中存在细粒夹层时,热成像图像能反映非均匀流场中热流峰面在分层界面处出现的"收束"现象;层状介质中的细颗粒夹层可导致热流锋面沿水流方向迁移速率下降、热量沿垂直于水流方向运移范围加大、温度分布更均匀。同时对于热源持续输入的系统,热成像温度在前期能较好的反映层状介质对水流运动的影响。将热示踪与HYDRUS模型相结合可较好地用于反演介质水热运移参数,反演所得的饱和导水率估值随粒径的减小显著降低,纵向热弥散度随粒径的减小而增大,而横向热弥散度变化趋势与之相反;纵横弥散比变化范围在10~120之间,且纵横弥散比随粒径减小而逐渐增大。对细砂饱和导水率的估计不足及热量损失是造成水流通量估计误差的主要原因,在模拟模型中增加细砂层测点数量可显著降低水流通量的估计误差。本研究可为非均质介质中水热迁移过程模拟与参数反演提供相应的方法。

Hydraulic and thermal parameters are the most important factors to simulate water and heat trans- port processes in porous media quantitatively. In this work, a sand box was packed with layered silicon sands and a laboratory experiment was carried out with steady-state flow based on heat tracing theory. Ther- mocouple and thermal imager were used to monitor inner and surface temperature of porous media at the same time. Temperature from both method were used to estimate parameters, such as saturated hydranlic conductivities, longitudinal and transverse thermal dispersivities at difierent locations of the sand box via the inverse model of HYDRUS-2D software. The results indicate that thermal images could shoot the behav- ior of water flow in different sand layers through the early process of transportation. The migration rate of heat front slow down within fine sand layer, so the heat in fine sand layer could expand to a larger spaee and temperature become more uniform along the way perpendicular to the flow direction. The results also show that temperatures information from both inner and surface is a feasible way to estimate hydraulic and thermal parameters of layered porous media at saturated condition. The estimated saturated hydraulic conduc,- tivity and longitudinal thermal dispersivity enlarged with the increasing of diameter of silicon sand, while the estimated transverse thermal dispersivity possessed the opposite tendency. The ratio of longitudinal to transverse thermal dispersivity ranged from 10 to 120. Besides, the heat loss during heat tracing test and relatively lower estimated value of saturated hydraulic conductivity in the fine sand layer were the major rea- sons ior the underestimating of water flux. With the number of observation node in fine sand layer increas- ing, the relative error between estimated and measured water flux declined from 37.7% to 20.7%. The re- suh from this experiment provides a method for estimation of hydraulic parameters of heterogeneous porous media using heat tracing.