Research For Petrographic Analysis Method On Complex Lithology Of S Block

S block is a part of the West African passive continental margin basin. It has abundant oil and gas. Its exploration potential is enormous. In this area, the rock of Albian formation is often complex and strong heterogeneous. It has brought great difficulties to us in the oil and gas exploration and development. In the reservoir of complex lithology, the different lithologies often have different skeleton parameter values, porosity and permeability. The wrong lithology will give the reservoir improper parameters, lead to get incorrect computed result. Therefore, the rock must be divided to different type and different reservoir evaluation models must be built. The lithology identification has become a main problem we face. In this area, the work as follows has been done with the complex lithology:First, a basic understanding of reservoir is obtained through relations research; Then, the conventional lithology identification methods were tried; Finally, electrofacies-lithofacies knowledge were applied to research.Through the research on petrophysical properties, we know mixed lithology is the difficulty and focus in this area. Through the use of conventional lithology identification method, mixed lithology identification was found with multiple solutions, the effect is not ideal. Therefore, we introduce electrofacies-lithofacies knowledge to identify lithology. Lithofacies in this article is not strict geological sense, but the major categories of rocks which have difference petrophysical properties.Electrofacies-lithofacies identification method mainly to complete the following aspects:1. The characteristics of the logging data were extracted. By studying the different on the logging curve characteristics and conventional crossplot analysis, we selected those curves which reflect lithology well for feature extraction. The feature extraction method includes the standard deviation of the standardized method and principal component analysis. In the lithofacies identification, the paper proposes a different principle of selecting the main ingredient, and it has been proved by analog data.2.Method of building lithofacies knowledge base by electrofacies is adopted. Generally, rocks is described according to its geological properties while logging data describes rocks according to its petrophysical properties. Because of the purpose and observation scale differences of these two properties,it is not adoptable to identify rocks by petrophysics which is defined by geology. This paper uniforms the differences and achieves combination of logging data and geological data.Based on petrophysical properties of rocks, electrofacies are obtained by K-means clustering and are given corresponding geological properties by correlating them to core data. Weight factor of principle component logs is explored and readjusted during clustering. Cost function and statistical analysis methods are introduced while obtaining electrofacies. Discussion focused on how to determine appropriate numbers of electrofacies avoids the blindness by experience in the past. Similarities and dissimilarities between different electrofacies are compared after obtaining them.3.Two different discrimination models are developed on the basis of electrofacies-lithofacies knowledge base, according to principles of Bayesian and Fisher discrimination method respectively. Contrast analysis of results from these two models shows Fisher model is more accurate. Finally, Fisher discrimination model is adopted and applied to other wells in the same area. Besides, the contrast analysis reveals discrimination model based on electrofacies is more accurate than that of lithofacies.