Prediction of Penetration Rate for PDC Bits Using Indices of Rock Drillability, Cuttings Removal, and Bit Wear

Abstract

Predicting rate of penetration (ROP) has gained considerable interest in the drilling industry because it is the most-effective way to improve the efficiency of drilling and reduce the operating costs. One way to enhance the drilling performance is to optimize the drilling parameters using real-time data. The optimization of the drilling parameters stands on the fact that drilling parameters are interrelated; that is, corrections in one factor affect all the others, positively or negatively. Analysis of the available models in the literature showed that they did not take into account all factors, and therefore, they might underestimate the ROP. To improve the accuracy of predicting the bit efficiency, a new ROP model is developed to preplan and lower the drilling costs. This approach introduces three parts of the process that were developed to describe the challenge of predicting ROP: aggressiveness or drillability, hole cleaning, and cutters wear, which are interrelated to each other. The approach discusses each process individually, and then the influence of all three factors on ROP is assessed. Taking into account the drilling parameters and formation properties, ROP1 is estimated by use a new equation. Then, lifting the produced cutting to the surface and evaluating how that affects the bit performance is proposed in the second part of the process (hole cleaning). Finally, wear index is introduced in the third part (wear condition) to predict the reduction of ROP2 caused by cutter/rock friction. The approach serves and could be considered as a baseline to identify all factors that can affect the bit performance. The developed model equations are applied to estimate ROP in three vertical oil wells with different bit sizes and lithology descriptions in Libya. The results indicate that the driven model provides an effective tool to predict the bit performance. The results are found in good agreement with the actual ROP values and achieve an enhancement of approximately 40% as compared to the previous models.