Fractured biological limestone reservoirs are characterized by structural complexity, strong inhomogeneity, and dual pore networks, resulting in a lack of technical methods for fracture identification for biological limestone, so it is a big problem to recognize and evaluate the fracture development of this type of reservoir. Wang Xuzhuang oilfield was taken as an example, based on various data such as core data, computed tomography(CT), injecting thin sections conventional logging curves and imaging logging(FMI) imaging, fracture reservoir prediction models were constructed. The "micro" fracture was identified through the comparison of the core photo, thin section and CT, then the response characteristics of dual laterolog for different production fractures were summarized and micro-fractures could be found in the biological limestone phase of the research section according to the large difference in the amplitude of the electrical resistivity. A back propagation(BP) artificial neural network was constructed based on nonlinear operations, the single-well fractured reservoirs were delineated. The comprehensive probabilistic index approach was adopted to realize the prediction of the distribution of fractured reservoirs in the whole region. The characteristics of oil and gas production at the fracture development and non-development places were compared and summarized. The research results provide theoretical basis for the research on the relationship between fracture development and oil and gas production.