Researches On Interpretation Technologies Of Batch Horizons Based On Constraint

In seismic exploration techniques, seismic data interpretation is a very important part. The result of the data interpretation directly affects the effectiveness of seismic exploration. Among the many aspects of data interpretation, the horizon and fault interpretations are the core part. Most traditional seismic interpretation techniques use manual two-dimension method. However, as the increasing of the seismic data, the effectiveness of traditional manual methods cannot meet the production requirements. Thus, automatic or batch interpretation algorithms have attracted more and more attentionMultiple horizons interpretation which is based on up and down horizons constraint is the actual production assists common requirement. However, how to comprehensive utilization of multilayer plane information, at the same time consider the effect of fault to the horizon tracking is still a difficult problem in horizon interpretation. This article carried on the thorough research to this difficult problem, then to get the following main results:(1)A research of the current automatic tracking technologies is implemented for follow-up study. Current automatic tracking technologies include dip-driven, horizon patches, artificial neural network, image edge detection and so on. These algorithms consider the characteristic properties of seismic data, and then analyze the properties from different aspects to pick horizons automatically.(2)Presently, most approaches can only identify the single horizon at one time. In this thesis, the automatic tracking algorithm tracks multiple horizons synchronously. This algorithm consider the information of constraint surfaces and the restraint among horizons which to be picked, and then pick up the appropriate samples in adjacent seismic traces by spread method and it has a good efficiency and accuracy.(3)Consider the relative knowledge about geomorphology, this thesis proposed an algorithm which converts the problem of horizon matching across faults to energy minimization. The problem consists of two parts: horizon-pairs and combining horizon-pairs. Then design an energy function to describe the similarity of the properties of the horizons on both sides of the fault, and the difference between the ideal fault throw and the computed fault throw. Finally, get the optimal matching configuration by optimizing the energy function.