Abstract
Statistical analysis methods have been widely used in all industries. In well logs analyses, they have been used from the very beginning to predict petrophysical parameters such as permeability and porosity or to generate synthetic curves such as density or sonic logs. Initially, logs were generated as simple functions of other measurements. Then, as a result of the popularisation of algorithms such as the k-nearest neighbours (k-NN) or artificial neural networks (ANN), logs were created based on other logs. In this study, various industry and general scientific programmes were used for statistical data analysis, treating the well logs data as individual data sets, obtaining very convergent results. The methods developed for processing well logs data, such as Multi-Resolution Graph-Based Clustering (MRGBC), as well as algorithms commonly used in statistical analysis such as Kohonen self-organising maps (SOM), k-NN, and ANN were applied. The use of the aforementioned statis-tical methods allows for the electrofacies determination and prediction of an Rt log based on the other recorded well logs. Correct determination of Rt in resistivity measurements made with the Dual Laterolog tool in the conditions of the Groningen effect is often problematic. The applied calculation methods allow for the correct estimation of Rt in the tested well.
Highlights
Dual Laterolog tools were designed for such conditions, but they fail in certain situations [12,13]
As a result of the cluster analysis using the Multi-Resolution Graph-Based Clustering (MRGBC) method, six electrofacies were separated in Well-4 and propagated to the remaining wells in the considered depth intervals
The research described in the article aimed to find out to what extent recently popular The such research described the article aimed to find out to whatcluster extent recently methods, as data mining,inmachine learning, neural networks, analysis,popular can be methods, such as data mining, machine learning, neural networks, cluster be helpful in the interpretation of well log data
Summary
Statistical methods applied in the well logging analysis were used to predict parameters that could not be determined directly from measurements [5,6]. High resistivity deep resistivity (LLD) readings in the reservoir zone isolated from the top by highresistivity formations are known as the Groningen effect [14]. Additional measurements can be planned to eliminate the Groningen effect It is worse with archival well logging data, where we can find many falsified LLD logs from the DLL tools. In the age of widespread digitisation, it is popular to build digital models of the reservoir [21,22] In such situations, one has to be careful when entering well logging data into reservoir models in carbonate formations [23]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
