RESERVOIR MODELING UNDER MULTI-SOURCE CONDITIONS-A CASE STUDY FROM YONG 935-936 BLOCK IN YANJIA OILFIELD

Due to the incomplete data and the different research methods, the prediction results of reservoir modeling will produce greater uncertainty. How to effectively reduce and evaluate the uncertainty has become a core issue in the work of reservoir description.

In this study, the ultra-low porosity and ultra-low permeability reservoir in Yong935-936 block of Dongying Sag was taken as an example to carry out uncertainty modeling. In view of the existence of multiple provenances in the study area, the discontinuity of the sand body in the partition boundary was solved by designing a locally variable azimuth variogram model. At the same time, the boundary of the fan body is delineated according to the geological knowledge, and the percentage of sand and mudstone under the lower limit of the physical properties of different effective reservoirs obtained by statistics in the range of the different fan body is used as the conditional data in the modeling, so as to reduce the uncertainty in the modeling process. Use the full factor experiment to analyze the sensitivity of the parameters that affect the reserve calculation, clarify the significance of each influencing factor, construct a multiple regression model between the reserve calculation and the significant influencing factor through the response surface experiment, and finally combine the Monte Carlo stochastic simulation method to obtain the distribution of cumulative probability reserves is to predict the 3P reserves in the study area and test the validity of the multiple regression model.