Research And Implementation Of Seismic Horizon Tracking Visualization System

The oil and gas exploration is dependent on seismic data heavily,and the accuracy and efficiency of seismic data processing and interpretation have a direct effect on the efficiency of oil and gas exploration.Seismic horizon tracking is one of the most basic work of seismic data interpretation.Generally,most of traditional seismic horizon tracking is done by manual work.But with the deepening exploration of oil and gas,the type and quantity of seismic data is increasing continuously,and artificial interpretation can hardly meet the demand of production.In addition,most of the interpretation systems are displayed on two-dimension space.So it is necessary to further study seismic horizon tracing visualization systems for massive 3D seismic data.This paper discusses the existing methods of automatic seismic horizon tracking and 3D seismic visualization,and researches the approaches of seismic data denoising and enhancement.According to the experiments of comparing the median filter,mean filter,Wiener filter and Gauss filter in seismic data denoising,median filter performed better among all the methods and was used to denoise seismic data.Then Laplasse operator was used to enhance the denoised seismic data in order to make seismic events have a better continuity,which is helpful for automatic horizon tracking.This work also analysed the correlation tracking algorithm,artificial neural network algorithm,image edge detection tracking algorithm and seismic DNA tracking algorithm to find the shortcomings of existing algorithms,especially DNA tracking algorithm,and optimized this algorithm by introducing correlation algorithm and clustering method to seismic DNA algorithm in order to improve the continuity and accuracy of tracked seismic horizon and finally implemented automatic seismic horizon tracking based on modified seismic DNA algorithm.After realizing automatic seismic horizon tracking,this work also conducted research on 3D visualization methods.The speed and performance of triangular mesh generation using three algorithms: growth algorithm,partition algorithm and incremental insertion algorithm were evaluated,and incremental insertion algorithm was chosen as the best algorithm for triangular mesh generation.Finally,a seismic horizon tracking visualization system was designed and implemented based on improved seismic DNA algorithm,and the framework of the interpretation system was designed in detail,realizing the functions of seismic data management,seismic horizon tracking and seismic horizon visualization.The proposed seismic horizon tracking visualization system has achieved good performance in practical applications,which could improve the accuracy and speed of seismic data interpretation,demonstrating that the system has high practicability.