A petrophysical modeling-guided method for predicting parameters of low-permeability reservoirs
[Backgroud]Accurately predicting reservoir parameters is significant for characterizing subsurface reser-voirs,establishing gas accumulation patterns,releasing production capacity,and understanding fluid migration.The tra-ditional approaches based on core measurement or mathematical-petrophysical modeling are limited by the strong multi-plicity of solutions and low accuracy of elastic parameters inversion results,making it difficult to meet the demands of modern exploration.[Objective and Methods]To more effectively predict reservoir parameters,this study proposed a petrophysical modeling-guided method for predicting parameters of low-permeability reservoirs.With the convolutional neural network(CNN)as a deep learning framework,the proposed method can predict water saturation,clay content,and porosity based on actual seismic data.Additionally,considering insufficient labeled data,the petrophysical model-ing combined with the random perturbation of elastic parameters was adopted to generate high-quality training samples,thus effectively expanding the size of sample data.[Results and Conclusions]The theoretical model tests demonstrate that:(1)This method can effectively predict the spatial distributions of parameters of low-permeability reservoirs in the case of low sensitivities of reservoir parameters to petrophysics.(2)Compared to data-driven deep learning,this method can yield high-accuracy predicted results of reservoir parameters based on merely a few log data.As substantiated by ex-ploration in the Dongfang block of the Yinggehai Basin,the proposed method facilitates the optimization of well deploy-ment,guiding the achievement of significant exploration breakthroughs and reserve discovery in the low-permeability areas of the basin.
deep learningreservoir parameter predictionconstruction of labeled datalow-permeability reservoirpet-rophysical modeling
国家科技期刊平台
NSTL
万方数据
