The Numerical Simulation Research Of Tunnel Seismic Prediction

Currently, in the construction of tunnels for domestic railway, highway, water and electricity infrastructure, tunnel seismic prediction (TSP) advanced forecasting technology is used based on geological predictions; however, current literature focuses on the actual application effect analysis, and based on numerical simulation, numerical modeling analysis, wave field separation, velocity analysis and imaging aspects of this research are very limited. Therefore, to improve tunnel geological prediction accuracy, carrying out tunnel earthquake advanced prediction technology based on numerical simulation research is of significant importance.The primary purpose of this paper are the following:the use of ANSYS to simulate 2D seismic waves of a tunnel seismic wave field, with research on the reflection and refraction wave field characteristics while seismic waves propagate in the face of the rock and surrounding rock media; research on the effective reflection wave extraction methods and effect evaluation of the TSP system; research on velocity analysis and migration imaging technology of the TSP system through numerical simulation; research on wave field propagation and recording features of layered and non-layered media (such as karst caves); and finally, by comparing the value simulation results with the experimental results, analysis of the sensor depth impacts of the received signal characteristics.This paper concludes with common geological hazards, such as coal seam gas, faults, high ground stress and karst caves, with each geological defect described in detail. The author adds pictures about tunnel construction with the use of TSP prediction in which he has participated and emphatically analyzes a fault and karst cave, which are typical geological features.The author gives a detailed description of the simulation steps of the wave equation finite-element modeling by stating the wave equations of the longitudinal and transverse waves. Based on ANSYS, the numerical simulation of the tunnel seismic wave field is realized, and the results of the numerical simulation are shown to be feasible and effective by the time record and snap shot plots of the wave fields. With respect to the layered model, we study the low speed model by numerical simulation, such as the vertical monolayer, the multilayer,30 degrees of inclination and 60 degrees of inclination, and analyze the propagation characteristics of the tunnel seismic wave field. For karst cave models, we use an empty model with radiuses of 3 m and 6 m by numerical simulation. Additionally, the propagation characteristics of the wave fields are also studied. The propagation of the tunnel seismic wave fields is analyzed based on the combination of the time record and snap shot plots of the wave fields. These models correspond to the TSP forecast of the geological engineering hazards, such as the fault and the karst cave. We can lay a foundation for the abstract of a reflective layer, wave field separation and migration.Wave field separation is studied based on wave field separation principles and tunnel seismic prediction methods laying a foundation for the methods and technology of P-S wave separation. We study P-S extraction and reflection in the TSPwin program as well as the basic properties and time windowing effects of the Radon transform. We research the reflection travel curves with multiple velocities and angles, and then, we obtain a variety of forms for the hodographs. We also study the response relationship of the thickness and plane wave and analyze the parameter selection of the discrete ?-p transform in the TSPwin program. For P-S wave spatial direction filtering, we can introduce spatial direction filtering from the direction feature of the waves and study spatial direction filtering in the TSPwin program. By letting the filter angle of the reflected wave be 90 degrees, the interference wave can be effectively eliminated, and the P and S waves can be separated. Comprehensive studies show that Radon transforms can separate reflected waves well. From the models of 90 degrees,60 degrees and 30 degrees, we can see that the effect of separation is best in the upright case, and the worst is 30 degrees of inclination. A good effect of separation is observed because the time-distance relation of the upright model and the Radon transform are both linear. When the model is declining, the time-distance relation is nonlinear, but the Radon transform is linear, so the effect of separation is not good. For the karst cave model, after the ?-p transform, there is a strong interference wave, but when the diameter of the karst cave is close to or greater than the first Fresnel zone, the karst cave is close to the reflector, significantly improving the separation efficiency; for karst cave models, the cloud image of the velocity analysis can provide more accurate positioning.Based on previous statements, from the research of the time-distance relation of the detector and each shot in the tunnel seismic prediction and from the principles of geometric seismology, we can see that the seismic signal fired by the shot reflects through the interface. The geological information of the ellipsoid is from the gun point to the reception point in the time recording carried out by the detector. Considering the characteristics of the tunnel seismic prediction, the diffraction stack is used in the offset calculation, and we can let the TSP data migrate images from time domains to spatial domains. The amplitude value is obtained from the corresponding time recording by calculating the different time-distance relations from the shot to the receiving point. The maximum is obtained by superposition. We can determine the position of the reflecting interface, and at the same time, the migration stack makes the diffraction wave converge to the real position of the diffraction spot. Accompanied with homing turning waves, the true form of the concave interface is recovered. While establishing the meshing, for tunnel seismic prediction, the default 1-m mesh spacing can meet the requirement. The offset is a relatively simple step after completing the velocity analysis. After completing the offset, the reflecting interface is extracted according to the amplitude and covering.By analyzing the wave field of the source buried in the loose and normal circles and the time recording of four different buried circumstances of the detectors, we process the time recording using the TSPwin program and obtain the corresponding reflecting interface. Then, we comparatively analyze the detector embedded in the loose circle with engineering examples. Through the analysis of the simulation and engineering data, we can see that the reflecting interface position processed by the TSPwin program is behind the actual position.Application of the front section of the research results in a practical geological engineering forecast in advance. By comparing the prediction results and the excavation results, which both have good compliance, it is shown that tunnel geological advance forecasting is effective according to the parameter setting in this paper.