Horizon auto-tracking technology based on the waveform similarity for the coalmining area' seismic data

Full 3D interpretation must be based on really full 3D horizons interpretation. In this paper, we used mathematical model of the similarity principle which express the characterization of the waveform similarity of seismic data, and we analyzed the effect of model parameters on the waveform similarity recognition, then we applied this method to auto-track the reflected wave of coalseam in the 3D seismic exploration of Liliangdian mining area. The results show that the waveform refers not only a conventional amplitude waveform, but also refers to the waveform of seismic attributes such as phase, frequency and variance etc. The principle of waveform selection is to highlight the consistency of the reflected wave of coal seam and obtain the most effective horizons information of the reflected wave. During the calculation of waveform similarity, there are some key parameters, such as the appropriate scope, the time dip and the reasonable time window and time dip, all of these is closely related to the reflection wave of coal seam in the study area. The correlation threshold value is the important parameters during the process of controlling horizon auto-tracking. The correlation threshold value is lower, and then the blank area of horizon auto-tracking is smaller. In the opposite case, the higher correlation threshold value, the larger blank area of horizon auto-tracking. The iterative loop in the horizon interpretation process is established as a key point for automatic tracking. The correlation threshold value is considered as characterization values which control the difficult degree of seismic waves auto-tracking. In the research area of Liliangdian coalmine, High correlation threshold value for horizon auto-tracking is firstly selected and the good result is set as new seed point. The correlation threshold value is adjusted for the unsatisfactory area in the follow tracking; finally, the horizon auto-tracking is gradual improved by stepwise cycle.