State Key Laboratory of Coal Resources and Safe Mining,China University of Mining and Technology(Beijing),Beijing　 100083,China

In order to improve the seismic identification accuracy of small fault,an automatic fault recognition method based on support vector machine (SVM) is constructed. The fault identification model of SVM two classification was established by analyzing the characteristics of seismic attributes of tectonic and non-tectonic sites. Firstly,the basic principle and structure of the algorithm about SVM two classification were studied. It showed that SVM has the advan- tages of high accuracy in two classification problems. Then,a fault forward model was established to analyze the fault response characteristics of different seismic attributes,indicating that the fault distribution is related to the trend of the attribute value. By analyzing the SVM and forward model,it showed that it is feasible to use the seismic attribute as learning modules to predict small faults. A set of attributes related to faults,such as variance and curvature,was extrac- ted from the post-stack seismic data. Four attributes with low correlation were determined by correlation and cluster analysis. Among total 606 data that consisted of the faults and attributes information from 14 drillings and 3 tunnels, 400 data were used as training samples and constructed SVM fault identification model. The authors used 206 data as test samples to identify faults. The correct rate is 98% . The SVM fault identification established by seismic attributes can effectively identify faults,reduce the influence of human subjective factors,and shorten the time of interpretation. The more uniform the drilling distribution,the more the number,the higher the interpretation accuracy.

SUN Zhenyu,PENG Suping,ZOU Guangui. Automatic identification of small faults based on SVM and seismic data[ J]. Journal of China Coal Society,2017,42(11):2945-2952.