Predicting Cracks In Coal Bed From 3D3C Seismic Data

Coal production is one of the most significant constituents of the development strategy for energy sources in China. The safety for coal mine production is an important prerequisite for coal supply. In recent years, there are two outstanding problems concerned with the safety in coal mines. One is gas explosion, and the other is water invasion and mine flooding. It is usually suggested that gas and water accumulation are associated with cracks(and/or fractures ). On the other hand, gas is a hazard but also a kind of resources of coal bed gas, known as the clean alternative energy sources of our country in the future. Effective exploration and production of coal bed gas are also closely related with the development situation of fracture zones. However, how to detect crack and fracture in mines remains unsolved at home and abroad. Thus it is a remarkble challenge to seismic exploration at present.More than ten years ago, it has been widely recognized in the community of coal exploration that 2D prospecting is limited. And it has been realized that the utility of P wave and converted waves could bring about new chance for detection of lithology and cracks. Therefore a great number of three-dimensional and three component (3D3C) seismic exploration has been made during the period of the“Tenth Five-Year Plan”, yielding a lot of valuable data in varied areas in China. Comparing with traditional seismic exploration of single component, the 3D3C can record simultaneously P and PS waves. The interpretation of PS wave is much more difficult and complicated than that of P wave, however. In last nearly ten years, it remains unsolved how to develop methods for real data processing to fully use PS wave that is the key of the 3D3C technique also. Currently processing and interpretation of real data are still confined to single-component reflective P wave.Coal seismic exploration has two prominent features. The coal bed is shallow which produces distinct reflective waves, thus original data with high signal/noise ratio and high excitation frequency can be obtained from 3D3C. The other is that the coal strata (including coal beds and upper and lower country rocks) is very thin with respect to seismic wavelength that different from oil exploration on land or sea, so it is often resulted as a reflection from simple interface. How to determine whether crack existing or not in coal beds? Whether the effects of cracks are appeared on seismic data or not and how to display them? These questions remain to be answered. In previous oil exploration, it was suggested to use splitting of S-wave reflection and/or difference of arrival time of splitting converted S waves to predict cracks in target layer. Such a method is primarily focused on target layer with nearly 100m thickness, and not applicable to reflection of thin bed of coal. In addition, it is not clear in theory whether the prevailing method using reflective P wave can be used to detect cracks in thin coal beds. Thus, it has been a debt whether the 3D3C method can be successfully applied to detecting cracks for coal mine production.Yao Chen theoretically proposed the dip CCP gathers and to use the apparent polarization anomalies of PS wave for detecting cracks. The most concerned problem in coal exploration is, however, whether cracks in thin coal beds can produce apparent polarization anomalies of PS wave. If the answer is positive, what is the relationship between their expressions and data processing of P and PS waves. Furthermore, the prediction of containing cracks or lacking could be tested by logging or drilling data? These problems are associated with whether our country is able to develop its own independent technology of detecting cracks in coal mine. It is also concerned with the development direction of the 3D3C in the future, as well as safety of coal production and exploration of new energy sources.In this paper, I firstly make analysis of the apparent polarization anomalies of PS wave for different azimuths, addressing the theoretical basis for using these anomalies to detect cracks. Then in conjunction with actual data handling, analysis and comparison of previous methods of data processing, the influence factors on the extraction of dip CCP gathers are studied as a focused topic. It is emphasized that the dispersion of conversion points on the interface would lower lateral resolution. Finally, by analysis of azimuth gathers, I compare the features of PS wave of two horizontal components to predict cracks in coal beds, and make comparison with the rock cores from wells that cracks detection has been carried out individually.This thesis includes three parts. In Part one, I compare reflections between lithological interface and crack layer by synthetic seismograms, demonstrating the differences of three-component reflection records for models with vertical and inclined cracks respectively. The purpose is to elucidate the apparent polarization anomalies of PS wave caused by PS wave splitting and how to use them to detect cracks in thin coal beds. Part two focuses on influence factors on dip CCP gather for 3D3C and complicated geological structure. I have explained why the previous efforts on 3D converted wave imaging are so difficult. The emphasis is focused on dip CCP stacking and imaging of PS wave in comparison with P wave. Part three indicates the advantage of using PS wave of dip CCP gather with multiple azimuths and multiple offsets that reveal the apparent polarization anomalies of PS depend on azimuth, indicating the single azimuth analysis is difficult to detect cracks.This work has made a comprehensive study in a chosen target area with both available 3D3C data and detailed well data. Based on comparison of reflections near the wells with and without cracks, it is found the there is indeed difference appearing as normal and abnormal polarization of PS-wave. This means that it is possible to find a new approach to detect cracks in thin coal beds from PS polarization anomaly.The research for detecting cracks in coal mine has important application value and prospects, in the background without experience at home and abroad, and it is carried out with great difficulties and risks. Many problems involved in this study have no relevant report from home and abroad.