Abstract: The drillability classification of the surrounding rock during the drilling process of coal mine gas drainage through the Mechanical Specific Energy(MSE) can provide a basis for the drilling rig to adjust the drilling parameters. Aiming at the problems of difficulty and low efficiency of manual layer identification in the process of gas drainage and drilling, a coal and rock drillability classification method based on MSE as the drillability evaluation index combined with Extreme Learning Machine(ELM) is proposed. A simulation model of PDC bit crushing rock was established by ABAQUS, and the changing law of drilling speed and MSE in the process of PDC bit crushing rock was studied from three aspects: material type, rotation speed and drilling pressure. At the same time, training data including drilling parameters and MSE were obtained. The ELM is used to learn the data such as drilling parameters and MSE. Finally, the classification accuracy under the two drillability classification indicators is compared. The results show that the accuracy rate of classification when the MSE is used as the index of drillability is over 90%, which is higher than the drilling parameters. The classification results can provide a theoretical basis for drilling rigs to adjust drilling parameters and realize adaptive drilling.