Abstract: In oil and gas field development, sedimentary micro-phase identification plays an important role in clarifying the sedimentary background and single sand body delineation. Only the bottom of Shan 1 section produces gas. For the analysis of sedimentary microfacies through multiple data intersections, relying solely on manual discrimination of sedimentary microfacies is a complex and error prone process, making it difficult to establish an accurate correspondence between sedimentary microfacies and logging data. Therefore, in order to make full use of the logging data and improve the efficiency of sediment microphase delineation, this paper proposes an automatic identification of sediment microphases based on Adaboost algorithm to provide a more accurate basis for sediment background and single sand body delineation for later gas field development. In the study, optimization and preprocessing of logging curves were carried out, and six feature parameters were extracted using mathematical statistical methods as the input set for training. The type of sedimentary microfacies was used as the output result label for training, and a total of 1210 groups are selected as training samples from the interpreted sedimentary microphase data, of which a total of about 968 groups of training samples are formed and 242 groups of test samples are formed. The results of the study show that the accuracy of the training results and test results of the application of the method reaches 96.45% and 90.4%, respectively, which can be verified that the method is better applied in Q area of Longdong gas field.