Formation pore pressure plays a vital role in oil drilling, and is an indispensable basic parameter for casing scheme design and mud specific gravity optimization in oil drilling. Formation pore pressures can be predicted through the data of seism, logging, and logging-while-drilling. However, due to the concealment and complexity of carbonate formations, as well as the inherent errors in the date of seism, logging, and mud loging, formation pore pressure is always difficult to predict accurately. Thus, a probabilistic method for predicting formation pore pressure is proposed to quantitatively describe the uncertainty of the pressure. Firstly, the method in this paper provides statistical properties of Eaton index and the normal compaction trend line distribution of random well depth. Then, with the Monte Carlo simulation method, the random number corresponding to the distribution features can be generated. And further the pore pressure sample set of any depth can be calculated. Finally a normal distribution is selected to fit the pore pressure sample set of any depth, and the cumulative probability distribution of pore pressure at any depth is derived. The pore pressure values with the cumulative probabilities of 0. 05 and 0. 95 at each depth point were selected and connected in series along the entire well section to obtain a formation pore pressure interval profile with a confidence of 90%. The case study shows that the method, integrating the logging and information recorded, obtains more accurate pore pressure prediction results, which provides a reference value for the uncertainty analysis of pore pressure in carbonate formation.
