Geological Modeling And Tracking Methods In I Oil Formation Of Nj₂ Member In Weizhou 11-1 Oilfield

Weizhou 11-1 Oilfield have been developed for a long time,which have a small number of wells and rich seismic data.However,the implementation of follow-up development plans will be seriously affected because of the less understanding of reservoir internal structure characteristics and the abnormal water-cut-regularity of development wells.Making full use of the limited dynamic and static data for fine geological research and building more delicate geological model that are consistent with the actual geological conditions are particularly critical in this situation.Morever,the internal structure and physical differences are increasingly complex in development so that the original geological model can't meet the needs of production development.In order to guide the oilfield production more effectively in future,it is urgent to carry out real-time tracking and constantly improve the geological model by integrating of the latest geological knowledge and production dynamic data into the model.In this paper,selecting Nj₂I oil formation reservoirs in the Weizhou 11-1 Oilfield as the study object,fine geological research?such as stratigraphic,sedimentary,interlayers and reservoir heterogeneity?is carried out with the data of core,logging,earthquake and production dynamic data.On the basis of this,according to the characteristics of mainly horizontal wells and seismic data in the work area,a geological model which can characterize reservoir heterogeneity is established by using the probability constraint method based on multiple trend integration with the limited dynamic and static data.In order to real-timely understand the law of reservoir physical properties,it is preferable to establish the fine geological model for Nj₂I oil formation reservoirs of the eastern area in the Weizhou 11-1 Oilfield.After the model accuracy test,aiming at the permeability with non-Gaussian distribution,a new model dynamic tracking method constructed upon normal transformation ensemble Kalman filter?NT-EnKF?is proposed that applies EnKF into the normal transformed vectors.Then,the updated vectors are switched back to the initial distribution after geological model updating by reversing the transformation,in order to maintain the non-Gaussian structure of priori model.Finally,the real-time updating of reservoir static parameters?porosity and permeability?is realized.The result of model tracking based on NT-EnKF shows that the proposed method is superior to traditional EnKF in terms of geological knowledge and production and dynamic data verification.The updated model based on the NT-EnKF not only inherits the advantages of the traditional EnKF,but also maintains the initial non-Gaussian characteristics.Furthermore,the updated results better reproduced the change of porosity and permeability and fundamentally solved the condition that the water cut is complicated and the fitting is difficult in the study area.It provides an important basis for the remaining oil recovery and reservoir prediction in the Weizhou 11-1 oilfield,and also provides an effective means for the tracking and development of complex oil and gas fields with non-Gaussian distributions.