Study Of Porosity Prediction In Deep Overpressure Environment

Porosity is an important parameter in reservoir description. Reservoir porosity is mainly influenced and controlled by sedimentary facies and diagenesis. Abnormal formation pressure is a common phenomenon in deep hydrocarbon-bearing basins,which is closely related to the generation,migration and accumulation of hydrocarbon reservoire.Researches show that overpressure is a common phenomenon in deep hydrocarbon-bearing basins of Dongying depression. The development of overpressure system of the Shahejie Formation was caused by high sedimentary rate, whereas both clay mineral dehydration and hydrocarbon generation result in dramatically rising of fluid pressure of overpressure system since the Late Oligocen. Overpressure raises the activity energy of diagenetic reactions.In this paper the reservoir porosity prediction in the overpressure environment has been studied. The method of TDI(time-depth index) is used to predict reservoir porosity. The relationship between seismic velocity and pressure, and the overpressure identification method have been analyzed and discussed. We get the interval velocity throughing logging constrained impedance inversion. Porosity, pressure, velocity and TDI values are statisticed, and the relationships between pressure and velocity, velocity and TDI values, porosity and TDI values have been established. The keys of using TDI to predict reservoir porosity lie in establishing the correlation model and recovering burial history curve including decompaction correction and denudation thickness. The overpressure factors have been utilized in the burial history curve recovery, and the accuracy of the results has been improved. Some factors such as compation, denudation, initial surface porosity and overpressure were considered in the TDI computation process, and some conclutions which can be used to model establishtion have been obtained. In this paper, the porosity inversion method bas been applied in order to compare with the TDI method. The selection of some parameters such as seismic attribute number and operate length has been analyzed. The inversion effect of multi-attribute analysis and neural network method was compared. We established the correlation model of the reservoir measured porosity with TDI in different sediment microfacies, such as turbidite sandstone, beach-bar sand and glutenite. And using the correlation model we predict the porosity of Sha-3 and Sha-4 members of Dongying depression.