Upscaling in many cases is essential to reduce the computational time but there is always a disparity between the upscaled model behavior and fine geological model behavior. Up to now there has been an immense effort to reach an upscaling method which is in the most similarity with the fine geological model.

This paper presents a novel implementation of synthesizing statistical tools and intelligent techniques. Statistical tools such as cluster analysis are widely used in order to process multivariate data. It is an integrated approach for petrophysical characterisation of the well data and it can be successfully applied globally to both clastic and carbonate reservoirs with use of logs and core data to define petrofacies. Artificial Neural Networks (ANN) mimics biological information processing mechanisms. They are typically designed to perform a nonlinear mapping from a set of inputs to a set of outputs.

In this paper we used artificial neural network to make modern algorithm for cluster analysis, then by use of this algorithm we make state-of-the-art software for cluster analysis. By this procedure not only computational time is reduced but also the resemblance of upscaled model behavior to the real reservoir geological model is improved.

The procedure and its application in synthesizing cluster analysis and Artificial Neural Networks is demonstrated by a case study.

You can access this article if you purchase or spend a download.