| Interwell reservoir heterogeneity was an important controlling factor affecting exploration and development. Comprehensive evaluation of reservoir heterogeneity, which takes reservoir prediction as the primary content, was one of the most important researching contents in reservoir description. In view of the rigor situation of hydrocarbon exploration in Huimin Depression, the paper try to reveal the characteristics of distribution of sedimentary facies, connectivity of sandbody, distribution of reservoir parameters and flowing property of the central uplift belt, the major oil and gas distributed area of Huimin Depression, by the detailed researching of reservoir prediction of Jishan Sandbody. Many theories, methods or technologies such as the specialist decision model, artificial neural network, fractal geometry, geostatistics and flow-unit theory, and so on were used in the researching. And this was believed to be great technical support for exploration and development of Huimin Depression.Many spatial information, such as the provenance,the sandbody's major extension axis, the ratio of sandbody's length, width and thickness were gained by regarding the thickness of sandstone as the regionalized geological variable and taking variogram as the main researching tool. Some parameters with geological meanings were extracted to characterize the facies model gained from drilling cores. Microfacies of all wells were acquired by incorporating the expertise and BP algorithm of ANN. On this basis, some factors that impact the specialist's decision-making, such as thickness, distance, vertical position, azimuth and correlation of neighboring wells, were describe by quantitative mathematical models. A new innovative decision-making model and algorithm was developed and be used to construct the 3D data volume of facies simulation. And, some 3D and 2D figures were extracted from this data volume to analyse the sedimentary characteristics of Jishan Sandbody. Sandbody's connectivity decision algorithm was set up based on the 3D facies data volume. By calibrating the logging data with core analysis, interpretation models for the reservoir parameters were established and used. It was proved that well logging curves and reservoir parameters matched the fractal structure through phase space reconstruction and rescaled range R/S analyses, so the paper designed and established a fractal interpolating algorithm by taking SRA as the key method, a restruction algorithm for 2D curves and 3D surfaces by taking IFS as the key tool. The data volume of predicted parameters was then set up and evaluated in the geological meanings. A new method named"slicing and merging"was proposed innovatively to perform reservoir subdividing and identifying simultaneously. This method was used in flow-unit interpreting, reservoir flowing property predicting and evaluating. A new parameter named Integrative Flowing Index (IFI) was originally proposed and used to characterize the integrative flowing property of reservoir. Also, interwell fractal interpolation algorithm was used to predict the reservoir flowing property. On the other hand, this prediction researching can also be regard as a processing of construction, application and improvement of regional geological knowledge library. Lastly, conditions of hydrocarbon reservoir-forming and distribution of remaining oil were analyzed based on above researching results. And this can be used to direct the oilfield's exploration and development.This dissertation supply a good idea and practical research method for reservoir prediction, which mainly based on well logging data, macro-controlled by strata correlation, 3D microfacies simulation and sandbody connectivity, and take reservoir parameters interpreting and its interwell fractal predicting, flow units subdividing and evaluation of reservoir flowing property as the fine researching contents. This may be very useful for reference in reservoir heterogeneity study.
|
PDF Full Text Request