Mathematical Modeling to Predict the Rate of Penetration (ROP) Using Genetic Programming

Rate of penetration (ROP) model is a mathematical relation between bit penetration rate and properties of formation, drilling fluid and drilling operation conditions. Due to relatively high cost of drilling operations, it is essential to develop an accurate prediction of the ROP to estimate the drilling time and costs. In this paper, a new model has been developed for estimation of ROP in one of Iranian oil fields by implementation genetic programming. In the developed model, ROP has been correlated with 11 effective parameters reported in drilling master log and sonic log including weight on bit, bit rotational speed, total nozzle area size, mud weight, mud yield point, fluid loss and sonic time. For the evaluation of the proposed model, statistical parameters including root-mean-square deviation (RMSD), squared correlation coefficient (R2) and average absolute relative deviation (AARD) were calculated. Real data verification indicated that the developed model is accurate for estimating ROP and can provide useful information when drilling operation is running. The values of squared correlation coefficient and root-mean-square deviation show the reliability of the model.