The Method to Predict ROP of PDC Bits Based on Fusion of Theoretical Model and Machine Learning

In order to improve the prediction accuracy of the rate of penetration (ROP) of polycrystalline diamond compact (PDC) bits and provide a basis for field engineers to guide drilling production, the energy parameters of downhole screw drilling tools, hydraulic rock breaking, and rotary impact drilling tools were comprehensively considered based on the Teale model. In addition, the concepts of PDC threshold weight on bit (WOB) and threshold torque were introduced, and the calculation methods of WOB and torque were corrected. The theoretical equation of ROP of composite specific energy was established. Based on the concept of ROP ratio, a prediction model of ROP based on deep fusion of theory and data was established. The results show that the model not only points out the correct theoretical direction but also comprehensively utilizes the advantages of data-driven learning, and it further improves the prediction accuracy of ROP of PDC bits. The prediction accuracy of the fusion method was verified by the actual drilling data in Shunbei, and the ROP can be greatly improved by the optimization analysis of the example. This method can provide an effective quantitative tool for optimizing drilling parameters, evaluating rock breaking effect of drill bits, and upgrading speed-up tools, which has important application value.