Method and device for predicting yield of fractured well of loose sandstone reservoir

Abstract

The invention discloses a method and device for predicting the yield of a fractured well of a loose sandstone reservoir, and the method comprises the steps: determining the effective thickness of a target loose sandstone reservoir, and obtaining a reservoir porosity index, a reservoir permeability index and a reservoir shale content based on collected reservoir related data; a rock mechanical model is constructed, reservoir rock mechanical parameters are obtained, and a reservoir brittleness index is calculated according to the reservoir rock mechanical parameters; constructing a crustal stress field model, obtaining the maximum horizontal principal stress and the minimum horizontal principal stress of the reservoir, and further obtaining a two-way horizontal crustal stress difference index; and constructing a yield prediction model based on the porosity index, the permeability index, the shale content, the brittleness index, the bidirectional horizontal crustal stress difference index and the collected yield data of the existing fractured well. According to the method, geology, rock mechanics, ground stress and production data are fused, the yield prediction model based on the neural network is applied, and the accuracy of yield prediction of the fractured well of the loose sandstone reservoir is obviously improved.

Images