Research On Reservoir Prediction And Velocity Modeling Of Igneous Rock In W And G Areas

Igneous rock,as the most abundant rock mass in the crust,has become more and more influential in oil and gas exploration in recent years.On the one hand,igneous rock as unconventional oil and gas reservoir is successively found in oil-bearing basins such as Bohai Bay,Junggar,Songliao,and Erenhot,showing great potential for igneous oil and gas reservoir.On the other hand,the in-depth exploration of ultra-deep complex oil and gas reservoir in the platform and basin area of the Tarim Basin requires more and more precise imaging of Ordovician limestone faulted solution bodies.However,igneous rocks have a strong shielding and absorption effect on seismic wave,which seriously affects the imaging accuracy of faulted solution bodies.In order to ascertain the favorable distribution range of granite and andesite reservoirs in the W area of the Erenhot Basin,and the influence of the Permian huge thick igneous rock in the Halahatang's G area of the Tarim Basin on the seismic imaging of Ordovician limestone fault solution bodies,the reservoir prediction of igneous rock in the W area and igneous rock velocity modeling in G-work area are studied.First of all,the geological,seismic,core,slice,and logging data of the W and G areas are analyzed.Based on this,the seismic facies and seismic attribute of the target zone in the study area are extracted,which makes us clear about the igneous rock characteristics of the target interval in the study area.The growth of faults in the W area provides favorable conditions for reservoir formation.Its igneous lithology is granite and andesite.And there is obvious vertical zoning in the granite buried hill reservoir,and fracture-porosity and fracture reservoirs are developed.Whereas,the Andesite primary pores and dissolution pores are developed,and the overflow facies is favorable reservoir facies.After that,new attributes with good correlation with the thickness of the well point are obtained through attribute fusion in the W area,and the igneous rock reservoir distribution is preliminarily predicted.There exists huge thick igneous rock in the Permian in G area.The lithology is dominated by tuff,dacite,and basalt.It erupted in three phases.The first phase of the seismic profile is a chaotic facies zone,with medium and strong amplitudes,medium and high frequencies.The second phase is a parallel phase zone with strong amplitude and mid-high frequency,and the third phase is a low frequency blank phase zone with low amplitude and low frequency.Secondly,a three-dimensional fault model that conforms to the structure of the study area is established for the Andesite reservoir in the W area.The sensitivity ofthe gamma curve and the wave impedance curve to lithology is used to perform geostatistical inversion to obtain a fine wave impedance body and the Andesite probability distribution body.Through analysis,the quadratic function relationship between wave impedance and porosity is fitted to obtain the porosity body that reflects the physical properties of the igneous rock reservoir.Considering that the reservoir is related to many factors,the classification criteria are diverse,and the reservoir prediction is also essentially a classification problem,the high-precision wave impedance,porosity,and seismic data of the target interval are input into the support vector machine model trained based on the actual classification data.In the end,the most favorable reservoir area of the andesite is predicted,which agrees well with the logging results.So a more efficient and accurate prediction of the reservoir has been achieved.Finally,in order to solve the problem of velocity modeling of igneous rock anomaly,the seismic inversion velocity modeling based on db5 wavelet packet log curve reconstruction is implemented in G area.Acoustic logging curve can effectively reflect the underground geological information and the velocity of different rock formations,but they are easily affected by the surrounding environment of the wellbore.In addition,the information contained in the acoustic logging curve is limited,resulting in a reduction in resolution.The gamma logging curve has a clear response to the radioactive material contained in the underground rock layer,and the radioactivity of different rocks is different,which can be used to distinguish the lithology.Curve reconstruction can make full use of the advantages of logging curves such as gamma and acoustic wave.The db5 wavelet packet transform can accurately locate the main node information of the log curve.After that,the main nodes of each curve with sufficient resolution are fused with features to reconstruct a curve that can reflect the lithology and also have a certain correction effect on the original curve.And the reconstructed acoustic wave curve can avoid the interference of high-frequency noise and make the wave impedance curve more capable of distinguishing tuff,basalt and dacite.Later,the reconstructed acoustic wave curve is used in sparse pulse inversion.It is found that the reconstructed wave impedance profile of the igneous rock is clearer,and the velocity curve extracted along the well from the inversion velocity body is in good agreement with the original longitudinal wave velocity.And the relative error of the well point velocity verification in the target interval is within 2%,which meets the accuracy requirement for velocity modeling.