The Study On The Uncertainty Of The Reservoir Modeling

Uncertainty is a core problem of reservoir modeling study.In the case of imperfectexisting data and different research methods, Reservoir modeling results exist greatuncertainty, and can not reflect the reservoir geological condition comprehensively.In the caseof integrating the log data with the seismic data, using appropriate algorithm can control andreduce the uncertainty of the results effectively.This article mainly analyzes and studies the uncertainty in the reservoir modelingfollowing three aspects.1.Uncertainty of the reservoir modeling results caused by different algorithms. We studyof the uncertainty of the results obtained from Marked point process modeling algorithm,Sequential indicator modeling algorithm, and Multiple Point Statistics modeling algorithm，inresearched area I of the oil field in the eastern part of our country. In the case of200m and100m well spacing, we analyze the uncertainty of the three results.The results show that, in the case of200m well spacing, the uncertainty of the results iseven more serious. However, the uncertainty of Multiple Point Statistics modeling results isweaker than the other two algorithms.2.Uncertainty analysis and evaluation of probability reserves. In researched area I，wecarry out the statistical analysis of the probability reserves. The variances of the threealgorithms are0.68,1.23and0.45. It shows that the uncertainty of the probability reservesobtained by Multiple Point Statistics modeling algorithm is smallest. It also shows that theprobability reserves can be used to quantitatively characterize the uncertainty.3.Research of the uncertainty under the combination of sotf and hard data. In theresearched area II，we focus on the uncertainty of the results in two cases: logging data andlogging data+seismic data. And we make the uncertainty evaluation on the microfaciesmodeling results. Also, we put forward the method reducing the uncertainty in the reservoirmodeling.The results show that:(1) The uncertainty of the results obtained by logging data+seismic data, is smaller thanthe uncertainty of the results obtained by logging data. (2) Sources of uncertainty of the microfacies models:①the uncertainty caused by random seed;②the uncertainty caused by different algorithms;③the uncertainty caused by the degree of the seismic data control.(3) Under the combination of soft data and hard data, Multiple Point Statistics modeling algorithm is an effective method to reduce the uncertainty in reservoir modeling.