Abstract:

The characteristics of conventional logging curves under different fracture scales were analyzed, the logging curve classification methods of Class Ⅰ (well diameter expansion ratio >18%, deep resistivity < 20 Ω·m, interval transit time >260 μs/m, density <2.4 g/cm3), Class Ⅱ(well diameter slightly expanded, deep resistivity <30 Ω·m, interval transit time > 235 μs/m, density < 2.55 g/cm3), and Class Ⅲ (no significant changes in well diameter expansion, deep resistivity, interval transit time and density) fracture scales were established, and the fuzzy identification sample space was constructed on the basis of considering the fracture scale, and the membership degree function and equation of whether the reservoir capacity of the extended formation of Honghe oilfield was low or high production was determined. If the low yield degree of membership is greater than or equal to the high yield degree of membership, it is a low yield oil layer, otherwise it is a high yield oil layer.After the identification verification of 20 oil layers, the accuracy rate of this model is 85%, and a good identification effect has been achieved.

Key words: Honghe oilfield; , Yanchang Formation, oil reservoir, crack scale, fuzzy recognition, production capacity