A new comprehensive geological prediction method based on constrained inversion and integrated interpretation for water-bearing tunnel structures
A new comprehensive geological prediction method based on constrained inversion and integrated interpretation for water-bearing tunnel structures
复制标题
DOI：
10.1080/19648189.2016.1170731
复制
发表时间：
2016-04
影响因子：
2.1
通讯作者：
Shucai Li;Jie Song;Jianqing Zhang;Chuanwu Wang;B. Liu;Fangwen Liu;Shengming Ma;L. Nie
中科院分区：
文献类型：
--
作者：
Shucai Li;Jie Song;Jianqing Zhang;Chuanwu Wang;B. Liu;Fangwen Liu;Shengming Ma;L. Nie
超前地质预报和掌子面前方含水地质构造查明是保证隧道施工安全的有效途径。综合地质预测常用于获取掌子面前方不良地质情况，对各种预测方法进行综合对比分析，有助于减少解的重复性，提高可靠性。然而，在传统的综合预测系统中，每种方法的预测数据都是单独进行反演和处理的，因此实际上无法解决每个类别的检测结果多解的问题。为此，本文提出一种基于约束反演和综合解释的含水隧道结构综合地质预报新方法来解决这一问题。首先，优选四种预测方法，以满足含水结构高级检测的需要。它们是地震反射法、瞬变电磁法、3D电阻率法和探地雷达（GPR）法。然后，基于不等式和空间结构约束，提出了三维电阻率约束反演方法。不等式约束表征了探测区域的电阻率变化范围，其先验信息主要来自地质推断和岩心钻探分析。空间构造约束表征地质构造边界和形状，其先验信息主要来源于地震反射、探地雷达等一些较为有效的地质构造边界识别方法，也可以通过钻探揭示获得。电阻率约束反演减少预测多重解的作用主要体现在数据处理上，综合解释减少预测多重解的作用主要体现在结果解释过程中。同时，从弹性、电导率和介电常数的差异出发，总结了地震反射、电磁、三维电阻率、探地雷达等方法对含水地质构造的响应特点。因此，提出了含水结构的综合解释指南。最终通过约束反演与综合解释相结合，建立了新型综合地质预报技术体系。算例和工程应用表明，该方法能够充分利用不等式约束和空间结构的先验信息，结合对综合地质预测结果的综合解释，对于消除反演假异常、减少反演多重解、提高预测精度具有显着效果。
Advanced geological predictions and ascertainment of the water-bearing geological structures in front of the tunnel face are effective ways to ensure the safety of tunnel construction. Comprehensive geological predictions are often used to obtain adverse geological conditions in front of the tunnel face, and the comprehensive comparison analysis on various prediction methods can help to reduce multiplicity of solutions and improve reliability. However, in the conventional comprehensive prediction system, the prediction data for each method is separately inverted and processed, so the problem of multiple solutions in detection results for each category actually cannot be solved. Therefore a new comprehensive geological prediction method is proposed in this paper based on constrained inversion and integrated interpretation for water-bearing tunnel structures to solve the problem. Firstly, four preferred prediction methods are selected to meet the needs of the advanced detection of water-bearing structure. They are seismic reflection method, transient electromagnetic method, 3D resistivity method and ground penetrating radar (GPR) method. Then, a constrained inversion method of 3D resistivity is proposed based on the inequality and spatial structural constraints. Inequality constraint characterises the electrical resistivity variation range of the detection area, and its prior information is mainly from the geological inference and core-drilling analysis. Spatial structural constraint characterises the geological structure boundary and shape, its prior information is mainly derived from some more effective methods for geological tectonic boundary recognition such as seismic reflection and GPR methods, and it also can be derived from the disclosure by drilling exploration. The role of resistivity constrained inversion to reduce the multiple solutions of prediction is mainly reflected in data processing, and the role of integrated interpretation to reduce the multiple solutions of prediction is mainly reflected in the result interpretation process. Meanwhile, the responding characteristics of methods of seismic reflection, electromagnetism, 3D resistivity, GPR and others to the water-bearing geological structure are summarised from the differences of elasticity, electrical conductivity and dielectric constant. Therefore, an integrated interpretation guideline for water-bearing structures is proposed. Finally, a new comprehensive geological prediction technology system is established through combining the constrained inversion and the integrated interpretation. Numerical examples and engineering applications show that this method can take full advantage of the prior information of inequality constraints and spatial structures, and together with the integrated interpretation on the comprehensive geological prediction results, it has a significant effect in eliminating the false anomalies of inversion, reducing the multiple solutions of inversion and improving prediction accuracy.