The prediction of pore pressure through Eaton method requires a highly precise depth domain interval velocity model to make correct predictions

but the velocity from conventional interval velocity modeling methods cannot reach such an accuracy.So

an innovative prediction method is proposed

which is specifically suitable for the poorly compacted mudstone formation

characterized by abnormal high pressure and low velocity.Shot domain velocity coherence inversion method is proposed to establish the initial velocity model.The residual curvature analysis of common imaging point gather in angle domain(ADCIGS)is used for velocity convergence.The grid tomography based on layer constraints is used to make local modifications.These three methods are integrated as a new interval velocity prediction process

specific for abnormal high pressure zones.The new process is then applied to the modeling of depth domain interval velocity field and prediction of pore pressure in the working area M of Bohai Sea.As a result

the interval velocity model matches well with the acoustic logging velocity

and the pore pressure coefficient calculated by the new process is identical with the Palaeogene formation pressure measurements.Judging from these

we can see its accuracy is much higher than the conventional method.The realworld case study demonstrates that the depth domain interval velocity modeling can make highly accurate predictions on pore pressure

and thus are effective and feasible.These methods can improve the accuracy of pore pressure prediction while maintaining the efficiency of calculation

and thus can facilitate the safety of drilling engineering.