|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
ABSTRACT Facies classification is a crucial task which can improve the chances of success of a well significantly. The relevant classification algorithms take well logs as inputs and classify the formation into distinctive clusters or electrofacies. Integrating the electrofacies with core measurements can lead to an understanding of the geological facies. We develop a general-purpose workflow for unsupervised electrofacies classification, which takes well logs as inputs and can be used for different application scenarios. The clustering is performed using the Gaussian mixture model approach. The optimal number of clusters is automatically determined ensuring repeatable clustering results from multiple realizations of the classification workflow. The workflow was applied on field data from off-shore Norway. We observe high similarity in the resulting facies with the ones determined visually by the field geologist from core data, by comparing their permeability-porosity relationships. This new approach removes the user intervention in the workflow and provides a robust solution for automating the electrofacies classification processing. INTRODUCTION Facies classification is a key element in the evaluation of petrophysical formations and in reservoir characterization. Electrofacies are defined as clusters of similar log responses in a well or a set of wells and their combination with core measurements can lead to geological facies, which can represent series of petrophysical properties. There has been significant progress towards developing automated workflows for facies classification (Busch et al, 1987; Lim et al. 1997; Rabaute, 1998; Qi and Carr, 2005; Skalinski et al., 2006; Tang et al., 2011). There are three main challenges in electrofacies classification. First, the fact that most of the times there are no labeled data necessitates to use an unsupervised classification method. There are various unsupervised learning algorithms like the k-means (Lloyd, 1982) or the hierarchical clustering algorithm (Ward, 1963) to perform classification. However, these algorithms perform "hard" assignment of data points to clusters, in which each data point is associated uniquely with one cluster (Bishop, 2006) and they do not consider the fact that field data can have some uncertainty over the clusters they are assigned. Second, the optimal number of clusters is usually unknown and thus is required to be an input given by the user. Various approaches have been developed to avoid the user’s subjectivity in the choice of the optimal number of clusters and automate the process. Some of the most common ones are the Bayesian Information Criterion (Schwarz, 1978) and the Cross-Entropy Clustering (Tabor and Spurek, 2014), which however do not provide a universally robust solution. Finally, it is very common different realizations of the classification algorithm to give different clustering results, even if the input logs and the algorithm parameters are kept the same for all realizations. This is because each of the input parameters of the algorithm is initialized randomly for each realization and as a result the algorithm converges to a different value of loglikelihood and the clustering is different each time consequently.
Abstract Breakouts provide valuable information with respect to evaluation of maximum horizontal stress magnitude and also verification of the geomechanical model built for a field. Caliper and image logs are routinely used to identify borehole enlargement. However, these methods are limited in their applications in many instances. In addition, good quality image logs are not usually available in old fields. This led to the need for development of a new approach to identify borehole breakouts. Petrophysical logs are usually acquired in most of the drilled wells and some of them have good correlations with mechanical properties of the rock. In this paper, a new multi-variable workflow is proposed in order to identify the location of borehole breakouts along the wellbore in correlation with some of the petrophysical logs acquired using wireline or logging while drilling tools in addition to mud weight and in-situ vertical stress data. This approach employs number of data processing techniques including statistical classifiers, and wavelet de-noising to determine borehole intervals with maximum likelihood of enlargement. The results showed that analyzing de-noised petrophysical logs with Bayesian classifier enables identification of breakouts with a significant accuracy. This paper explains the methodology and presents the results in five study wells in a carbonate field. The study confirms the applicability and the generalization capability of the method in carbonate formations with a reasonable accuracy. Introduction The integrity of the wellbore plays an important role in petroleum operations including drilling, completion and production. Wellbore failure occurs principally through changes in the original stress state due to drilling the rock that concentrates stresses around a wellbore. If the new stresses exceed the rock strength, breakout will form around the borehole. Borehole enlargement may lead to difficulties removing cuttings and if severe can result in borehole collapse. The same factors contribute to the risk of solid production during the producing life of the well. To complete a successful wellbore stability analysis or sanding study, building a reliable geomechanical model is a basic requirement. Breakouts are valuable information to calibrate and verify a geomechanical model. In addition, they provide final borehole shape which is a critical factor in completion and production optimization. This study aims in identifying borehole breakout zones in carbonates from common petrophysical logs using data processing techniques such as wavelet decomposition, de-noising and statistical classifiers as an alternative for caliper and image logs which are associated with many limitations. Limitations of available methods Breakouts were first documented using 4-arm caliper data as zones that have consistent orientations in which one caliper pair indicates a borehole size that is greater than the bit size while the other caliper arm pair is in gauge (Wiprup 2001; Moos et al. 2007). Besides the caliper, several other downhole devices have been used for identifying borehole breakouts; including borehole optical televiewer (Gazaniol 1994) and both acoustic and electrical borehole imaging devices (Aoki et al. 1994; Van Oort et al. 1995; Tan et al. 1998). While caliper data are most often used in regional and field studies because they are widely available, acoustic imaging tools are considered the best devices for identifying breakouts and distinguishing them from other types of borehole elongations. Figure 1 illustrates a comparison between electrical and acoustic image logs' quality in breakout identification in an identical interval. As can be seen, the acoustic log displays very well-defined breakouts, whilst, electrical image logs do not show any visible breakout.
ABSTRACT Cement bond evaluation is a critical step in the early-life stages of newly drilled wells since it rules the way for obtaining useful information about wellbore integrity. Conventionally, this is carried out by means of a detailed interpretation of cased-hole sonic and ultrasonic log data. However, this standard approach can be highly time-consuming and challenging in long completion sections and when complex scenarios have to be handled in operative time. In this respect, oil companies have stored huge datasets for their wells, with quality-checked cased-hole acoustic logs and associated interpretations in terms of wellbore integrity. This paper deals with a novel, probabilistic data analytics approach aimed at obtaining a fast and robust cement bond facies classification. The latter is deemed able to automatically provide an exhaustive quantitative cement placement evaluation, hence avoiding time-consuming processes and possible subjectivity issues. The implemented methodology takes advantage of the Multi-Resolution Graph-based Clustering (MRGC) algorithm that gathers its knowledge by recognizing patterns in sonic and ultrasonic logs/maps from dozens of wells, including more than 500K meters of logged intervals. This allows the system to learn through experience how the log measurements are related to the common cement bond scenarios (e.g. good, partial, poor cementation, dry or wet microannulus, free pipe). The MRGC is then integrated in a Bayesian framework to obtain the probability of the cement bond facies, the most probable scenarios, and the associated uncertainty by means of entropy computation. In detail, an automated screening can be performed in newly drilled wells to detect possible problems of hydraulic sealing. The potentialities of the discussed method are demonstrated by real case applications consisting of cement log data collected from several blind-test wells. First, the probabilistic approach is used to predict the cement bond scenarios together with the uncertainties of their classification. Then, an unbiased evaluation of the results is performed. The successful outcomes coming from the final step of the workflow show how, with a statistically representative and good quality dataset, data analytics can efficiently mimic high-skill expert work in harsh circumstances and within a time-efficient template. In fact, this data-driven methodology takes few seconds to provide an exhaustive interpretation against, at least, one day with the conventional one.
Facies classification at the well location is generally based on sedimentological models, however, the extension of the classification of an individual well into the entire reservoir model is contingent on the calibration of a rock physics model that links rock and fluid properties with geophysical measurements such as seismic velocities and/or electromagnetic-derived resistivities. The goal of this work is to present a workflow to define a geologically consistent facies classification at the well location, to accurately reconstruct this classification using elastic and electrical properties, and to extend the classification to the 3D reservoir model. The initial classification at the well location is obtained using traditional statistical methods applied to computed rock properties such as mineralogical volumes, porosity, density and permeability. The facies reconstruction based on elastic/electrical properties is obtained using a Bayesian approach that combines rock physics with statistical models. The workflow is illustrated through the application to the Rock Springs Uplift field, Wyoming, which hosts several potential CO2 storage reservoirs.
Facies classification aims to assign a rock type or class to each location of a 3D reservoir model, based on the available rock and fluid properties. At the well location, the classification can be based on measured well log data, formation-evaluation computed curves, and core samples. However, far away from the well, most of these properties are not available and the classification must be derived from geophysical properties estimated from surface measurements, such as seismic and electromagnetic properties. Several methods for facies classification have been presented in literature (Doyen, 2007; Avseth et al., 2005; MacGregor, 2012). These methods generally differ from the mathematical approach (deterministic or statistical methods) and for the input data (core samples, well logs, or geophysical inverted attributes). The scale of core samples allows petrophysicists to generate a very detailed facies description; however, the extension of this classification to well log and to 3D reservoir models is difficult to achieve due to the lower resolution and data noise of well logs and surface geophysical measurements respectively. The aim of this work is to show that the integration of a rock physics model in the classification and the use of statistical methods for uncertainty quantification can overcome this limitation.