Research And Application Of Sequentially Conditional Simulation Methods

As a important part of geologic statistics, conditional simulation is a main developing trend of geologic statistics. Since the turning band method had been brought forward by professor Matheron, there are many scholars who apply themselves to research of conditional simulation, and invent some methods which are widely used. Recent years, conditional simulation are applied deeply in reservoir stochastic modeling. A lot of mature reservoir modeling software, which include Kriging estimation and conditional simulation module, have appeared.The reservoir description is always uncertain and the result of reservoir forecast has several results because the reconnoiter information is incomplete. Conditional simulation, which overcomes the smooth effect of Kriging estimation, not only reappears the relative structure of reservoir attribute, but also makes the known data to be conditional. People always use the stochastic simulation methods to model and forecast the reservoir because conditional simulation can satisfy the description and analysis of uncertain character of reservoirConditional simulation is divided into error simulation and sequential simulation. Sequential simulation is a method that combines the sequential route and Kriging estimation. It is the route of sequential simulation that seeks the ccdf of every grid point sequentially by the stochastic path and obtains the value from the ccdf. Sequential Gaussian simulation (SGS) and sequential indicator simulation (SIS) are very familiar. SGS is used for continuous data which meet Gaussian distribution. SIS is used for the discrete data and the continuous data which has been scattered.This paper applies sequential simulation in petrophysical modeling of a certain area, and discusses the applicability of SGS and SIS which can be used for Gaussian distribution data and discrete data. This paper also researches the randomicity of the conditional simulation method and coherence between the simulation data and the original data, and states the applicability of sequential simulation. The conditional simulation can reflect the extremum which can not reflected by Kfiging method. At last, the sequential simulation is used into the reservoir modeling and provides multi results which can describe the actual geological condition.