Font Size: a A A

Study On Application Of Seismic Pattern Recognition Of River - Phase Reservoirs Under Geological Information Constraints

Posted on:2016-04-11Degree:MasterType:Thesis
Country:ChinaCandidate:X XuFull Text:PDF
GTID:2270330467999737Subject:Geological engineering
Abstract/Summary:PDF Full Text Request
The fluvial face reservoir in Bohai area, the geological condition is complex, the reservoir thickness is thin and changeable, the connectivity is poor, so to develop such fields is difficult, facing various challenges. The traditional reservoir geological study methods have been unable to meet the current development needs, especially for sand bodies superimposed types in a small layer. The traditional method to research the sand bodies superimposed types is through the seismic response characteristics, but affected by the seismic data resolution, there is no one-to-one relationship between sand bodies superimposes types and seismic response, so we have to consider using other methods to research the reservoir sand bodies superimposed characteristics.In this thesis, based on the seismic attributes extraction and optimization, we use neural network method to study the sand bodies superimposed characteristics. Firstly, summarize the sand bodies superimposed features existing in fluvial face reservoir in Bohai area, build seismic forward models, extract seismic attributes, use seismic attributes optimization method to select sensitive attribute, classify the sand bodies superimposed types through the neural network method. The destination to do the models research is to study how many types of sand bodies superimposed features can be classified through neural network method based on seismic attributes. In the end, the study concludes that all the sand bodies superimposed types in the work area divided into6kinds of patterns.Then, the achievements in the study of the models are used in the actual data of Bohai area, predict the sand bodies superimposed patterns and the pattern recognition has achieved good results, prove the scheme applied in the paper to predict the sand bodies superimpose mode is feasible, and the scheme present in this paper has certain guiding significance to the actual data.
Keywords/Search Tags:Seismic attributes, Sand bodies superimposed, neural network, reservoirprediction
PDF Full Text Request
Related items