The Study Of Vertical Well Segmental Fracturing Optimization In Tight And Thin Interbed Reservoir

Thin interbed tight reservoirs have the characteristics of multiple vertical layers,thin single layers,large differences in physical properties between layers,and low reserve abundance.Usually,people use large-scale fracturing vertical wells or fracturing horizontal well techniques to realize reasonable development.Many serious problems occurred during the development of this kind of reservoirs including serious inter-layer interference,uneven use of small layers,large reserves losses,etc.In order to realize the uniform use of reservoirs,reduce inter-layer interference,and increase oil recovery rate,it is of great significance to rationalize the fracturing of such reservoirs.Based on the geological features and development status of the target block,this paper compares and analyzes the drawbacks of fracturing horizontal well development and large-scale general fracturing development of vertical wells,and finally decides to choose segmental fracturing vertical wells development.Meanwhile,this paper establishes a method for strata segmental division and crack parameters optimization.On the basis of summarizing principles of intervals division in thin interbed tight reservoir,a quantifiable strata segmental division method is established by synthesizing the entropy weight method,the fuzzy comprehensive evaluation and the optimal segmentation methods,to both take the static geological factors and dynamic development factors of reservoir into account,and avoid artificial intervention.The concept of loss function is proposed to quickly screen the optimal segmentation scheme.On the basis of stratification,the reservoir factors and fracture factors affecting the productivity of vertical wells are sequenced through orthogonal tests,which are reservoir permeability> fracture half length> reservoir thickness> fracture conductivity> crack width.At the same time,in order to reduce the large number of sample calculations in traditional vertical well fracture parameter optimization methods,a production capacity prediction model based on BP artificial neural network is proposed.Through the trained neural network,a large number of sample production capacities are predicted and a variety of fracture parameter maps are plotted to achieve the optimal fracture parameters,and finally,the goal of crack parameters optimization can be achieved.The proposed vertical well segment fracturing optimization method was applied to a thin interbed tight reservoir in Shengli Oilfield,and the development effects of 7 kinds of staged fracturing methods were compared.The results show that the error rate for the BP neural network established in this paper is as low as 1.5%,and the relative average error is only 6.47%,which indicate the accuracy of this model.With the development time of 1642 days as the time node,comparing the cumulative oil production under different staged fracturing methods,when the number of segments reaches 5,the vertical well production increased by 1.367 times,and compared with the optimal 2 times sements,the production degree was increased by 1.24%.At the same time,compared with the general fracturing,each layers’ average recovery reached the highest value of 0.42% at the optimal 5-stage fracturing.