Publications

Fracture characterization has been conducted in the Salawati Field at Kepala Burung block of Papua, Indonesia. This field has several carbonate facies reservoirs. The production of this field is believed has been controlled by fracture system. The carbonate rock characterization is quite complex, because of the complexity of various matrix, pore system, and also consider of chemical reaction produced from fluid interaction in interior wall of their pores space which make their wave propagation system becoming more complex. This carbonate complexity requires special treatment to precisely characterize the reservoir. To make strategy for characterizing the reservoir fracture in this field, it was proposed to integrate all of the information from core, well log and seismic attributes using artificial intelligence method. In thispaper, the latest technology for carbonate complex reservoir characterization using hybrid seismic rock physics, statistic and artificial neural network. This methodology is able to integrate a huge size of various datasets to produce “coherence correlation” between input data and their target. The data set consist of core data (i.e.: lithology, lithofacies, fracture intensity, fracture width, porosity), well log data (i.e. gamma ray, density, Sw, porosity, resistivity etc.), multi-attributes either pre-stack or post-stack of a different vintages of 2D seismic lines and seismic rock physics. The whole of input data was trained together using natural workflow which is also combined with statistical and artificial neural network. Afterwards they are used to predict several reservoir parameters. This method was applied in Salawati Basin field to predict the lateral lithofacies, fault and fracture, both fracture and matrix porosity distribution. In addition, the whole process of reservoir parameter prediction is done by using natural algorithm based on lithofacies prediction result, therefore the lithofacies is the first task which should be done before characterizing the other properties of reservoir. By using this approach, it can produce high accuracy on the reservoir parameter prediction. The accuracy of testing process shows that predicted reservoir parameter about 90 percent match with their reservoir parameter in the existing wells. Keywords: Lithofacies, fracture, carbonate, artificial neural network