The Application of Multistage Geometric Analysis Method in Natural Fracture Identification

Abstract Imaging logging can intuitively recognize fractures around borehole walls. However, expensive cost of imaging logging limits its popularization, so little data of imaging logging can be used. How to use conventional logging data is an important item in fracture recognition. Conventional acoustic logging has sensitive response to cavities and fractures. Fractal geometry variable scale analysis (R/S) in combination with acoustic logging data to calculate fractal dimensions of target layers is a new method. Fractal dimension based on R/S analysis can reflect the variability of acoustic logging data, namely reservoir vertical heterogeneity. The stronger vertical heterogeneity is and the better fractures develop, the higher fractal dimension is in fractured low-permeability sandstone reservoirs. So fractal dimension is used to forecast fracture development in oil-bearing formations of Y1, Y2 and Y3 which are low-permeability sandstone reservoirs in X area of Western oilfield in Ordos Basin. Besides a significant discussion about fracture development depend on quantitative analysis of the results from field testing. Dynamic analysis, field outcrops and coring well evaluation show the forecast results have a good agreement with the actual situation. The research results have certain significant guidance to fracture description. Introduction Development practices of low permeability oilfield in both domestic and overseas show that natural fractures play an important role in the practical development result of low permeability reservoirs[1]. So the research of natural fractures has become one of the key contents of reservoirs evaluation and forecast, and also one of the urgent needs of the oilfield effective development. Now there are lots of methods for fracture recognition and description[2]. But it is often difficult to effectively identify and predict the distribution of fractures because of the restrictions of data types and quantities in a specific oil-bearing block. Conventional logging data are the most among the existing data in X area of Western oilfield. Conventional logging data to establish fracture logging response mechanism models to recognize fracture distribution can be used? Through consulting a large number of literatures, there are examples which use fractal geometry variable scale analysis (R/S) in combination with conventional logging data to recognize fractures relative development at home and abroad [3].So based on the characteristics of low and ultra-low permeability reservoirs of Yanchang Formation in the Ordos Basin, ractal geometry variable scale analysis is used to make a tentative forecast of fractures development of Y1,Y2 and Y3 layers, the main target layers in X area of Western oilfield. Besides fractal dimension to conduct a comprehensive classification evaluation at Y1,Y2 and Y3 layers is also can be used. And the results of evaluation are nearly identical to actual productions which have some vital guidance for effective development of oil -bearings.