ROP modeling chronology, sensitivity analyses, and field data comparisons

Rate of penetration (ROP), the rate at which a drill bit breaks the rock underneath to deepen the borehole, modeling and measuring is widely used in industry to monitor drilling performance, optimize drilling parameters, detect abnormal pressures, and to improve drilling efficiency. The objective of this project is to run various simulations and models with field data in order to investigate the relationship amongst the parameters that influence rate of penetration and the limitations and advantages of each model. This paper analyzes six models: Bingham’s, Bourgoyne & Young’s, Winters, Warren, and Onyia’s, Hareland’s drag bit, Hareland’s roller bit, and Motahhari’s. An analysis of the models with respect to changes in lithology and with respect to changes in formation is included as an initial check for the models. As expected, the analysis done by formation yielded better results, with an improvement of roughly 5% for each model. When the data sets for wells drilled with drag bits were run for the drag bit models and two other extensive models for comparison, an interesting result occurred. The least amount of errors was always achieved by a non-drag bit model, but Motahhari’s model, a drag bit model, always gave the closest physical interpretation. Using a non-bit specific model, however, may lead to a better initial planning, as the non-drag bit models averaged outputted values closer in magnitude to the real data. This paper provides good practices on how to choose which model to use. As a general assessment, Motahhari’s model should be used for drag bits, and Winters, Warren, and Onyia’s for roller bits. Using other models is dependent on availability of data, well complexity, and desire to expand on design or confirm calculations.