Machine Learning-Based Probabilistic Lithofacies Prediction from Conventional Well Logs: A Case from the Umiat Oil Field of Alaska

Abstract

A good understanding of different rock types and their distribution is critical to locate oil and gas accumulations in the subsurface. Traditionally, rock core samples are used to directly determine the exact rock facies and what geological environments might be present. Core samples are often expensive to recover and, therefore, not always available for each well. Wireline logs provide a cheaper alternative to core samples, but they do not distinguish between various rock facies alone. This problem can be overcome by integrating limited core data with largely available wireline log data with machine learning. Here, we presented an application of machine learning in rock facies predictions based on limited core data from the Umiat Oil Field of Alaska. First, we identified five sandstone reservoir facies within the Lower Grandstand Member using core samples and mineralogical data available for the Umiat 18 well. Next, we applied machine learning algorithms (ascendant hierarchical clustering, self-organizing maps, artificial neural network, and multi-resolution graph-based clustering) to available wireline log data to build our models trained with core-driven information. We found that self-organizing maps provided the best result among other techniques for facies predictions. We used the best self-organizing maps scheme for predicting similar reservoir facies in nearby uncored wells—Umiat 23H and SeaBee-1. We validated our facies prediction results for these wells with observed seismic data.

Hydraulic Flow Unit Classification and Prediction Using Machine Learning Techniques: A Case Study from the Nam Con Son Basin, Offshore Vietnam

“…Neurons are then placed at the nodes in a lattice. This converts multi-dimensional data into a 1D and 2D discrete map [37]. The SOM clustering method uses the following steps:…”

Modeling Permeability Using Advanced White-Box Machine Learning Technique: Application to a Heterogeneous Carbonate Reservoir

“…Petrophysical rock typing, generally defined as grouping the reservoir rocks according to similar petrophysical and flow characteristics, has a wide variety of applications in drilling (e.g., prediction of high mud-loss intervals), − production (e.g., potential production zones, locating perforations, diversion system design in acidizing, and prediction of high-injectivity zones), reservoir studies (net-pay cutoff definition), , representative sample selections for special core analysis (SCAL) tests, − permeability prediction in uncored intervals as one of the essential applications, ,− and defining saturation functions for static and dynamic reservoir models. , Therefore, petrophysical rock typing is crucial for reservoir modeling, operation, and field development. ,− It is divided into two categories of petrophysical static rock typing (PSRT) and petrophysical dynamic rock typing (PDRT) based on similar static ( P c , S wc , etc.) or dynamic (related to fluid flow behavior).…”

“…Artificial intelligence (AI), an essential part of the engineering toolkit in recent decades, has been used to solve various environmental and engineering problems. ,− Machine learning (ML) is a subfield of AI that encompasses a variety of data processing techniques, including classification, regression, and clustering . Supervised and unsupervised techniques are the two broad divisions of ML. , Because of the ability of ML algorithms to recognize patterns and provide valuable predictions, the use of data-driven/ML algorithms has gained much attention in the energy, oil, and gas industry. ,,− Data-driven and ML techniques have been used when sufficient core data or, similar to most previous studies, a combination of core data and well-logging/seismic data are available to build predictive models. ,− The principal use of data-driven strategies and ML algorithms in petrophysics is rock typing and permeability predictions. ,− Most studies in the literature used different variations of support vector machine (SVM) and artificial neural network (ANN) to classify the reservoir into homogeneous clusters and then predict the permeability of each cluster.…”

“…12 , 13 Therefore, petrophysical rock typing is crucial for reservoir modeling, operation, and field development. 1 , 14 − 16 It is divided into two categories of petrophysical static rock typing (PSRT) and petrophysical dynamic rock typing (PDRT) based on similar static ( P c , S wc , etc.) or dynamic (related to fluid flow behavior).…”

Visualization and Semantic Labeling of Mood States Based on Time-Series Features of Eye Gaze and Facial Expressions by Unsupervised Learning

“…Compared with DL algorithms that require a vast of data, one important advantage for SOMs is to conduct steady learning with relatively lower computational resources and calculation costs. Recent research examples of clustering, visualization, recognition, classification, and analyses using SOMs comprise medical system applications [ 28 , 29 , 30 , 31 , 32 ], social infrastructure maintenance [ 33 , 34 , 35 , 36 , 37 , 38 ], consumer products and services [ 39 , 40 , 41 , 42 , 43 ], food and smart farming [ 44 , 45 , 46 ], and recycling and environmental applications [ 47 , 48 , 49 , 50 , 51 , 52 , 53 ]. We employed SOMs and their variants for the task of classification and visualization of mood states.…”