【作者】 尹小康；

【导师】 王绪本；

【作者基本信息】 成都理工大学 ， 固体地球物理学， 2010， 硕士

【摘要】 本文主要针对大地电磁在铁路长大隧道应用中遇到的实际问题,进行探讨研究,分析了静态效应和地形影响的原理,讨论了静态校正和地形校正方法,并通过模型正演研究总结了实际应用中的一些规律,并将这些规律应用到实际资料处理中,提高了资料解释的可信度;同时将钻孔、地质等先验信息转换为电性资料,进行电性参数约束反演研究,提高了大地电磁方法的分辨率。论文首先重点对静态效应问题及其校正方法进行了研究,分析了静态效应的产生机制,总结了静态效应的几个典型的特征:浅部的小规模的电性不均匀体在大地电磁测深过程中会产生静态效应;静态效应只影响电场数据的观测,而不影响磁场数据的观测;在双对数坐标系中,视电阻率曲线会沿视电阻率轴上下整体偏移,不改变曲线形态,而相位曲线不受影响;静态效应容易与异常混淆。考虑了多种方法进行静态校正,通过模型正演研究,主要对五点(七点)滤波、二维中值滤波和自适应汉宁滤波等数值分析方法和曲线平移进行了静校正效果分析,讨论了各种校正方法的优缺点,提出采用多种静态校正方法组合方式,并通过实例对静校正的效果进行了分析评价。然后对起伏地形模拟理论进行了论述,并对起伏地形视电阻率进行了定义,通过建立斜坡地形模型、不同背景电阻率的正、负地形模型和复杂地形含高阻模型进行了正演模拟研究,得出对于TE模式:在负地形处和台阶底部,视电阻率值偏小;在正地形处和台阶顶部,视电阻率值偏大;且起伏落差越大、电阻率值越小,畸变越大。对于TM模式:地形畸变主要表现在正、负地形处。在负地形处视电阻率值偏大;在正地形处视电阻率值偏小;且起伏落差越大、电阻率值越小,畸变越大。对比值法地形校正进行了介绍和模型研究。将地质、地球物理等前期成果资料转换为大地电磁资料处理所需要的电阻率分布先验地电模型,在OCCAM反演算法的目标函数中加入先验地电模型作用项——即迭代过程中在已知电性分布的区域不断的向先验地电模型靠拢,形成了利用先验信息的电性参数约束反演算法,详细推导了目标函数的迭代公式,并对最终形成了电性参数约束反演模块进行了介绍。通过模型对比来说明约束反演的有效性。以渝利线方斗山隧道为例,主要是利用钻孔、测井等信息转换为电性参数约束反演所需要的电性约束文件,进行约束反演对比研究,通过详细对比约束反演前后的效果,说明了电性参数约束反演可以在一定程度上提高大地电磁方法在铁路长大隧道中勘探的分辨率。最后,将以上讨论的方法应用到南涪线梓里隧道的实际处理中,对于梓里隧道针对该工区位于构造发育,地下情况复杂等特点,对野外原始数据进行去“飞点”和地形校正等操作,采用空间滤波和汉宁滤波等方法进行了静态效应的校正,同时加入已知的地质先验信息进行约束反演,得到的预处理结果效果明显改善,最终的梓里隧道成果解释与钻孔具有较好的对应性。更多还原

【Abstract】 The paper paid attention to the problems which suffered from magnetotelluric (MT) application in large and long tunnels of railway and were studied. Static effects were discussed by two-dimension finite element forward. And some laws in practical application from forward research to the model whose application in data processing can enhance credibility in data interpretation. At the same time, posterior information such as borehole and geology etc was transformed to electricity data for inversion with electrical parameters restraint, and increased MT resolution.In the paper, Emphasis on static effects and their calibration research, analysis of their generation and their typical characters were summarized such as shallow small-scale electricity inhomogeneous body in MT sounding producing static effects; static effects having effects only on electric field data observation but not on magnetic data observation; in log-log coordinate apparent resistivity curve along apparent resistivity axis up and down offset without changing the pattern and with no effects on phase curve; easy confusion in static effects and anomaly. Curve translation and numerical analysis were considered for static calibration. By forward to models, five-point (seven times) filter, two-dimension median filtering and adaptive Hanning filtering and so on were used for static calibration. And advantages and shortcomings of various methods of calibration were discussed and effect of calibration was analyzed by practical examples.Then constructing various two-dimension geo-electric models was used to discuss terrain influence and vertical and horizontal resolution in MT application. To the terrain influence, the following laws were summarized. Including there no good reflection on the bottom interface of low-resistance anomaly; in the terrain of ridge they well reflected center position of anomaly and the terrain had little effects on the top interface of anomaly; in the terrain of valley, center position of anomaly can up and down offset, the greater of radian, the better reflection on low-resistance anomaly on the whole. Research of vertical and horizontal resolution found that MT method can very well determine top interface of the low-resistance anomaly in most upward side, but can not determine top interface of each low-resistance, that is, there is bad vertical resolution in the method. For horizontal resolution, in traverse, when there is some separation in low-resistance anomaly distance, the method can reflect horizontal variation of electricity structure very well. And MT application was discussed further by complex structured model.Geological and geophysical former data was transfromed to resistance distribution priori geo-electric model which is needed by MT data processing. Interaction item of priori geo-electric model was added into objective function in OCCAM inversion algorithm—during the iteration known electricity distribution continually closed to priori geo-electric model which came into being the inversion algorithm with the restraint of electricity parameters, iterative formula of objective function was deduced. And the final inversion module with the restraint of electricity parameters was introduced. Models comparison showed the availability of restraint inversion.Finally the methods which have been discussed above the paper were applied into practical processing in Zili tunnel of Nanpei line and Fandoushan tunnel of Yuli line. For Zili tunnel, according to the work area located in structural zone and its complex subsurface picture and so on, removing the“flying spot”of raw data, spatial filtering and Hanning filtering etc were used for static calibration. And obvious improved pre-processing results were obtained. Also final results interpretation of Zili tunnel corresponded to the results of borehole very well. For Fangdoushan tunnel, the information of borehole and logging etc used mainly was transformed to the electricity restraint file which was required by inversion with restraint of electricity parameters for restraint inversion comparison. And detailed comparison showed the inversion with restraint of electricity parameters can improve the resolution of MT exploration in large and long tunnel of railway in some degree.更多还原