Automatic Management of Rate of Penetration in Heterogeneous Formation Rocks

Abstract

Rate of penetration (ROP) is dependent on several factors and it is essential to properly manage and control this in order to save cost of drilling operations. Drillbotics test skid is an automated machine that is expected to drill as vertically as possible into a rock formation of varying compositions. This thesis demonstrates the design approach to controlling ROP by managing the effects of circulation system, WOB, RPM, bit torque and the reaction time of the controller in order to achieve an optimal drilling operation. Also covered are the detailed steps based on the design guidelines for the drillbotics 2016 competition. The results obtained during testing indicated high potential of controlling ROP automatically during drilling operations. The second part of the thesis focused on multiple regression techniques which were used in predicting the ROP of a well by using the coefficients obtained from a near-by well on the same block. The analysis was based on four wells, a pair on two blocks respectively. The technique considered the combined effects of drilling parameters, MSE and D-exponent independently in order to predict the ROP of the nearby well. The results displayed both, the actual ROP and the modelled ROP plots for comparison.