人工气候箱环境下膨胀土裂隙演化定量分析与预测模型

Expansive soils,with remarkable expansion, shrinkage and cracks, result in complicated and long-term damages to projects during engineerring constructions, which are deepened by extreme droughts and floods due to global abnormal climate changes.Under the background of expansive soil slopes and protection projects,this research subject is expected to, explore the quantitative evaluation method for cracks assesment and method for depth measuring,reveal the laws of crack evolution in plane and depth,create related theories and forecasting models.Firstly, cyclic wetting and drying generated by artificial climate box simulating actual atmosphere circumstance, is used to prompt and control the evolution of cracks of expansive soil in the box.Secondly, the information of cracks comprised in images by camera is drawn out to evaluate the features of cracks quantitatively, and the method for crack depth measurement is studied according to three-dimensional reconstruction from layered-images of cracks.Based on the test data with the methods, the relationships between crack evolution of shallow in-situ expansive soil and surface temperature,moisture and initial water content,etc., and the relationships between plane and depth characters of cracks ,are explored.Thirdly,the empirical model of crack depth forecasting and stochastic model of crack growth based on Markov Model are created for in-situ expansive soil, and used for the cracks of numerial simulation and prediction subjected to local climate. The achievements are benificial to enrich the theoretical research of unsaturated expansive soil and promote the theory and design level of damage assessment and slope surface protection for expansive soil slopes.

膨胀土的胀缩性与裂隙性所诱发的工程灾害具有反复性和长期性，并因全球气候异常所致极端干旱和雨涝而加剧。课题以膨胀土边坡稳定与防护工程为背景，探索裂隙演化定量评价和深度测试方法，揭示原状膨胀土裂隙平面演化和深度扩展规律，建立相关理论及预测模型。首先通过人工气候箱模拟干湿循环大气环境，促控箱内膨胀土裂隙发育，采用摄影图像形态学方法定量分析裂隙特征，通过裂隙分层图像三维重建手段研究深度测试方法，据此获取多组合试验数据，定量揭示浅层膨胀土裂隙演化规律及与表面温、湿度及持续时间、土体初始含水率等参数的关系，探求裂隙平面与深度特征的关联性，进而基于膨胀土裂隙的平面与深度特征关系建立其开展深度经验预测模型，基于Markov模型建立适用于膨胀土裂隙表面扩展规律的随机过程模型，并实现对气候条件下裂隙演化的数值模拟和预测。成果对丰富膨胀土的非饱和理论研究、推进膨胀土边坡灾害评估及坡面防护理论与设计具有积极意义。

本项目以膨胀土裂隙为研究对象，采用数码摄影及三维激光扫描成像、CT扫描等方法记录裂隙演化过程，通过改进裂隙图像定量化处理方法并编制相应计算程序，然后应用上述方法对大尺寸人工气候箱试验获得的压实膨胀土在不同初始条件及人工气候环境下的裂隙发育规律进行分析研究。项目总体上达到了预期目标，取得的代表性研究成果如下:.（1）改进了裂隙图像定量化处理方法，获得更为清晰可靠的裂隙二值图像，更准确地进行膨胀土裂隙长度、宽度、条数、方向、裂隙率等特征参数的提取与统计；基于MATLAB编制了相关程序，实现裂隙特征参数的定量化自动化处理。.（2）开发了一套人工气候箱试验系统，一套裂隙扫描测试系统，一套图像处理软件。其中人工气候箱系统可实现对箱内土样或地表土湿度、温度的组合控制，可供尺寸为50cm见方的土样进行试验；裂隙三维激光扫描系统可实现在土样的三维扫描与成像，初步实现对裂隙三维特征的测量；土体裂隙图像自动处理软件可实现对裂隙扫描结果的三维成像及裂隙特征分析。.（3）研究发现了膨胀土表面裂隙发育特征与初始压实度、含水率及环境温度、湿度之间的关系。湿干循环时第二次循环对膨胀土裂隙发育影响最显著；压实膨胀土的初始含水率越高、压实度越大，失水收缩后的裂隙率、长度与宽度等特征参数越大；环境温度越高，初期裂隙发育更快，但后期总体发育程度反而较小；湿度越高，初期裂隙发育越慢，但后期总体发育程度反而较高。这些裂隙发育规律可为相关工程防护与治理提供科学依据和技术支撑。.（4）初步发现压实膨胀土的含水率与颜色空间的R、G、B值存在较好的对应关系，深入研究后有可能提供一种简便可靠的含水率测试新方法。.本项目有关膨胀土裂隙发育的研究成果可为边坡、路基等膨胀土表面防护治理提供基本的科学依据及重要的技术支持，也具有较好的理论意义和较高的应用价值；摄影图像特征处理技术及裂隙三维扫描系统还可应用于墙面、地表等裂缝的监测，具有很好的应用前景。