Research On Prediction Methods For Unconventional Oil And Gas Sweetness In Gao 89 Well Block

For the low permeability reservoir,sweetness refers to the local favorable area under the overall background of poor property,low saturation and low abundance,it has important value in the unconventional reservoir exploration.We can first establish forward model based on the characteristics of target area reservoir,then study the corresponding relationship between the seismic response,seismic attributes and reservoir,finally extract sweetness attribute which can be expressed in a mathematical formula,that is the root of the ratio between reflection amplitude and instantaneous frequency.However the majority of unconventional oil and gas sweetness have the characteristics of the wide transverse distribution and the thin vertical thickness,which is also the most common type of sweetness.Therefore,aiming at the thin layer sweetness that have above characteristics,firstly we need to solve the problem of thin interbed identification,which can lead to a better prediction for the sweetness.The Beach-bar of Es4 in Dongying Sag work area is our research object reservoir of G89 work area,which is low permeability reservoirs,therefore it is part of the unconventional oil and gas reservoir,it has the geologic characteristics of deep burial depth of thin reservoir,poor physical property,low noise-signal ratio and low resolution of the initial seismic data.In this paper,we first make a research through reading a large number of domestic and foreign literatures,have a general understanding of the history of the unconventional oil and gas exploration and make a summary of research status of the prediction methods for unconventional oil and gas sweetness.Next we introduce the location and formation characteristics of the research area,make a detailed seismic interpretation for it,according to the geological features of research area,we establish a single channel seismic record,two-dimensional and three-dimensional sweetness model,take a forward model test,and then compare the recognition effect between the three instantaneous attributes and sweetness for sweet spot.Then we introduce sweetness attribute,illustrate the definition of sweetness and extract other ratio attributes which are similar to sweetness.Finally,there exist a difficult problem for the sweet spot prediction,that is,the resolution of the thin interbed is not high and it is difficult to identify top or bottom interface of sandbody,to address this problem,we bring in signal reconstruction algorithm based on compressed sensing,come up with inversion method based on the splitting Bregman iterative algorithm and orthogonal matching pursuit algorithm with compressed sensing,then try to apply these two methods in single channel,two-dimensional model test and actual data,it improves the resolution of thin interbed sweetness,which provides a good research foundation for the subsequent reservoir interpretation and evaluation.