pressure loss estimation of three-phase flow in inclined annuli for underbalanced drilling condition using artificial intelligence

underbalanced drilling as multiphase flow is done in oil drilling operation in low pressure reservoir or highly depleted mature reservoir. correct determination of the pressure loss of three phase fluids in drilling annulus is essential in determination of hydraulic horsepower requirements during drilling operations. in this paper the pressure loss of solid-gas-liquid three-phase fluids flow in inclined annulus was estimated using artificial neural network (ann). experimental data which are available in the literature were used for design of ann. pressure loss as output of ann, was estimated from five effective parameters as inputs of ann including gas and liquid superficial velocities, the inclination from horizontal, rate of penetration (rop), pipe rotation speed (rpm). the correlation coefficient between predicted and experimental value for train and test data is 0.998 and 0.997 respectively.the root mean square error (rms) and average absolute percent relative error (aape) for train data are 0.0082 and 2.77% and for test data, they are 0.0108 and 3.68 % respectively. the reliable results showed the high ability of artificial neural network for estimating pressure loss of three phase flow in annulus.