基于地基雷达的滑波形变监测与分析

发布时间：2018-08-13 09:41

【摘要】：滑坡是四川盆地周边较为普遍的典型地质灾害之一,它一方面受到地球本身构造运动等内因的作用、另一方面也受到水利设施和地下工程等人类活动的影响,给周边居民造成了严重的人员伤亡和经济损失。因此,有必要对其进行及时、准确的监测,以便进行应急管理并确定适当的预防措施；积极发展科学有效的新型滑坡监测途径与预警技术迫在眉睫。针对现有地面监测手段与空间遥感技术在空间分辨率、测量精度、气候条件制约以及及时性、有效性方面存在的诸多局限与不足,近年来新兴的地面合成孔径干涉雷达(地基雷达)在该领域取得长足发展,已充分展现出其巨大的应用潜力与重要的现实意义和经济价值。然而地基雷达所采用的短波段在实际监测过程中易受大气的影响。国内外不同学者提出了各自的解决办法,但是由于大气的复杂性,各种方法存在一定的局限性,目前还未建立有效的统一模型。针对此问题,本文在结合国内外学者的基础上提出了多模型大气改正算法,并用于实际滑坡监测与分析中。再者地基雷达采用斜距投影方式使得成像与常规的影像存在很大的区别,特别是在地形复杂的山区,使得图像解译非常困难。有鉴于此,本研究在视域分析的基础上对投影改正进行了改进研究。本文的具体工作如下：(1)为了更好地研究大气改正,参考国内外不同学者研制角反射器的基础上自主研发了三维移动平台。该平台具有高精度、易操作、稳定、体积小等优点。可模拟空间三维移动,验证测量仪器的精度。(2)提出了多模型大气改正模型。该模型从稳定点的选择、模型的选取、精度评定三个方面形成了自动一体化的流程。该算法从信噪比和相位稳定性出发选取有效的稳定点,在时域内根据稳定点的个数采用不同模型对比分析,并选择合适的模型,减少了由于人工干预造成的误差。并用于大气分析和滑坡监测实例中,取得了良好的效果。(3)基于可视域分析的投影改正。为了有效地消除投影改正中出现的错误,本研究在投影改正之前对获得的数据进行视域分析,去除雷达视角中的不可见区域,提高查找效率和可靠性。通过模拟研究表明该算法具有实际可操作性。(4)基于地基雷达的雅安硗碛水库滑坡监测与分析。硗碛水库周边滑坡严重,严重威胁人员和财产安全,本文利用地基雷达对其进行监测研究。本次监测采用本文提出的多模型改正算法有效地去除了区域大气影响。结合所获得的数据与实际调查,分析滑坡的时空演变特性,表明该处滑坡具有蠕变特性。研究表明该监测区域两处区域形变比较大,值得特别注意。通过理论和实例结果表明,本文提出的多模型改正算法具有自动一体化的优点,能减少由于人为因素造成的大气改正误差,在滑坡实际应用中表明具有良好的适用性。通过模拟研究得出基于可视分析的投影改正算法能够有效的提高投影改正的可靠性,具有实际应用可行性。
[Abstract]:Landslide is one of the typical geological hazards around Sichuan Basin. On the one hand, it is affected by the internal factors such as tectonic movement of the earth itself, on the other hand, it is also affected by human activities such as water conservancy facilities and underground engineering, which has caused serious casualties and economic losses to the surrounding residents. It is urgent to develop scientific and effective new monitoring approaches and early warning techniques for landslides. There are many problems in spatial resolution, measurement accuracy, climate constraints, timeliness and effectiveness of existing ground monitoring methods and spatial remote sensing techniques. In recent years, the ground-based synthetic aperture Interferometric Radar (SAR) has made great progress in this field, which has fully demonstrated its great application potential, important practical significance and economic value. However, the short-wave band used by ground-based radar is easily affected by the atmosphere in the actual monitoring process. In view of this problem, this paper proposes a multi-model atmospheric correction algorithm based on the combination of domestic and foreign scholars and applies it to the actual landslide monitoring and analysis. The oblique projection makes the imaging very different from the conventional image, especially in the mountainous area with complex terrain, which makes the image interpretation very difficult. In view of this, this study improves the projection correction on the basis of visual field analysis. The specific work of this paper is as follows: (1) In order to better study the atmospheric correction, the reference country. Based on the research and development of corner reflector by different scholars at home and abroad, a three-dimensional mobile platform has been developed independently. The platform has the advantages of high precision, easy operation, stability and small volume. The algorithm chooses effective stabilization points from signal-to-noise ratio and phase stability, compares and analyzes them with different models according to the number of stabilization points in time domain, and chooses appropriate models to reduce the error caused by manual intervention. (3) Projection correction based on visual field analysis. In order to effectively eliminate the errors in projection correction, this paper analyzes the obtained data in the visual field before projection correction, removes the invisible area in the radar view angle, and improves the search efficiency and reliability. Interoperability. (4) Landslide monitoring and analysis of the Ya'an Qiaoqi Reservoir based on ground-based radar. Landslides around the reservoir are serious, seriously threatening the safety of people and property. This paper uses ground-based radar to monitor and study the landslide. The multi-model correction algorithm proposed in this paper effectively eliminates the regional atmospheric impact. The data obtained and the actual investigation show that the landslide has the creep characteristics. The study shows that the deformation of the two areas in the monitoring area is relatively large, which deserves special attention. The atmospheric correction error caused by element has good applicability in the practical application of landslide. The simulation results show that the projection correction algorithm based on visual analysis can effectively improve the reliability of projection correction, and it has practical feasibility.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P642.22

【参考文献】