Machine Learning Based 1D Mechanical Earth Modeling and Wellbore Stability Analysis

Rock mechanical properties is essential for several geomechanical applications such as wellbore stability analysis, hydraulic fracturing design, and sand production management. These are often reliably determined from laboratory tests by using cores extracted from wells under simulated reservoir conditions. Unfortunately, most wells have limited core data. On the other hand, wells typically have log data, which can be used to extend the knowledge of core-based mechanical properties to the entire field. Core to log integration of rock mechanical properties and its interpretation is limited by our current understanding of rock physics. The gap is clearly evident where approximations such as empirical relationship between dynamic and static mechanical properties are used for rock failure estimation. This paper presents a hybrid framework that combines advances in digital rock physics (DRP) and machine learning (ML) to predict rock mechanical propertiy (e.g., Young's modulus) from rock mineralogy and texture to improve the accuracy of mechanical properties determined from log data. In this study, mineralogy, density, and porosity data are obtained from routine core analysis and rock mechanical property from triaxial compression tests. In our methodology, we utilized DRP models which were calibrated against core data and then generate rock mechanical property, for intervals for which triaxial measurements were not available. Mineralogy and texture data are used to create DRP models by numerically simulating rock-forming geological process including sedimentation, compaction, and cementation. Rock mechanical properties derived from DRP are used to enhance the set of training data for the ML algorithm to establish a correlation between rock mineralogy, texture, and mechanical property and construct the ML-based rock mechanical property model. The ML model generates Young's modulus predictions and are compared with the laboratory measurements. The predicted Young's modulus of rock models from the combined approach has a good agreement with the laboratory measurements. Two quantitative measures for estimation accuracy are calculated and examined including the correlation coefficient and the mean absolute percentage error. Cross-correlation plots between the Young's modulus predicted from the ML model and experimental results show high correlation coefficients and small error. The results of the study show that DRP model can be used to feed the ML model with reliable data so that the prediction accuracy can be improved. The results of this work will provide an avenue of learning from the formation lithology and using the knowledge to predict rock mechanical properties.

The demand for wearable devices that can detect anxiety and stress when driving is increasing. Recent studies have attempted to use multiple biosignals to detect driving stress. However, collecting multiple biosignals can be complex and is associated with numerous challenges. Determining the optimal biosignal for assessing driving stress can save lives. To the best of our knowledge, no study has investigated both longitudinal and transitional stress assessment using supervised and unsupervised ML techniques. Thus, this study hypothesizes that the optimal signal for assessing driving stress will consistently detect stress using supervised and unsupervised machine learning (ML) techniques. Two different approaches were used to assess driving stress: longitudinal (a combined repeated measurement of the same biosignals over three driving states) and transitional (switching from state to state such as city to highway driving). The longitudinal analysis did not involve a feature extraction phase while the transitional analysis involved a feature extraction phase. The longitudinal analysis consists of a novel interaction ensemble (INTENSE) that aggregates three unsupervised ML approaches: interaction principal component analysis, connectivity-based clustering, and K-means clustering. INTENSE was developed to uncover new knowledge by revealing the strongest correlation between the biosignal and driving stress marker. These three MLs each have their well-known and distinctive geometrical basis. Thus, the aggregation of their result would provide a more robust examination of the simultaneous non-causal associations between six biosignals: electrocardiogram (ECG), electromyogram, hand galvanic skin resistance, foot galvanic skin resistance, heart rate, respiration, and the driving stress marker. INTENSE indicates that ECG is highly correlated with the driving stress marker. The supervised ML algorithms confirmed that ECG is the most informative biosignal for detecting driving stress, with an overall accuracy of 75.02%.

The Orange Basin situated offshore South Africa has been subject to various studies as it is gas prone, but limited literature exists on the determination of fluid contacts. The availability of well logs, repeat formation tests (RFT), and drill stem test (DST) results make the Orange Basin an ideal example to carry out studies on the integration of these data to understand the fluid contacts in the Albian age gas bearing sandstone reservoirs. This paper compared pressure gradient results from repeat formation test, drill stem analysis, and well logs in the Orange Basin Offshore in South Africa, for the identification of fluid types and contacts in the Albian age sandstone reservoirs. In most of the wells, the interpreted water pressure gradient ranges from 0.42 to 0.45 psi/ft (0.98 to 1.04 g/cm3 equivalent). Results reveal the field water pressure gradient of 0.43 psi/ft (1.01 g/cm3). The combination of pressure data helps to improve the interpretation of conventional log data and provide useful information on the prediction of fluid contacts, as demonstrated in this study. Findings from this study provide insights during the phases of petroleum exploration and development to understand pressures that would improve drilling safety. Additionally, previously unclear fluid contacts from well logs are revealed and easily predicted with a better understanding of connectivity and isolation of reservoirs.

In drilling engineering, the mechanical properties of formation rocks are essential basic geological parameters. One of the most important tasks at the drilling site is to grasp the mechanical properties of the formation rocks timely and accurately. We take the readily available cuttings on site as the research object to test the compression characteristics. The experiment shows that the compressive stress of group cuttings with the uniaxial compressive strength of rock has an obvious correlation. The relationship model between compressive stress of cuttings compressed to 105% of original rock density and uniaxial compressive strength of original rock is established, and the results conform to an exponential function. When cuttings with different particle sizes (2-10 mm) are used, the coefficient values before the exponential term of the model are stable and the same, only the coefficient of the model is slightly different. The Group cuttings method can reduce or eliminate the disadvantages of discrete experimental results caused by a single cutting defect and individual differences. The new method takes cuttings returned from the drilling site as the test object, which is convenient and quick and can provide a way for the site to obtain the rock mechanics parameters of the drilled formation timely and accurately. Introduction Rock mechanics plays a significant role in the petroleum industry, especially in oil & gas drilling and development projects. The study of rock mechanics aims to know the geological engineering parameters such as rock mechanics properties of the reservoir, understand and master the change law of rock mechanics properties, and apply this law to drilling engineering (Qin, 2014). The work includes formation rock drillability analysis, reasonable bit selection, drilling parameter optimization, and wellbore stability analysis. It is one of the most critical tasks to grasp the mechanical properties of formation rocks in time and accurately (Lin, 2014).

Sand control decisions are often made based on a deterministically predicted Safe Drawdown Pressure (SDP) without proper regard to the amount of uncertainty associated with the value of SDP. These uncertainties can be large when planning a Drillstem Test (DST) for a deepwater exploration well. On one hand, predicting too low of a SDP for a DST can result in unnecessary sand control. On the other hand, predicting too high of a SDP can lead to sanding during the test, which can cause numerous problems and ultimately cost much more than a sand control installation. Thus, a probabilistic study is warranted to quantify the expected sanding risk through a SDP probability profile, which can be used to estimate the risked expected value of a decision to install sand control or not. In this study, data from 12 offset wells in nearby blocks from an Offshore Nigeria exploration well were reviewed. First, a regional geomechanical model (in-situ stress, pore pressure and rock mechanical properties) was calibrated and validated based on offset data. The calibrated geomechanical model was used with a sanding prediction model and was calibrated and validated based on offset DST and sand control data. Next, a statistical analysis of regional in-situ stress, sand pressure and rock properties for different sand intervals was performed to generate the necessary Probability Distribution Functions (PDF) and used as inputs to a SDP Monte Carlo simulation. And finally, a deterministic approach was also used to predict SDP. A comparison of Monte Carlo simulated SDP distribution and the deterministically predicted one from offset wells indicates consistent results. The SDP PDF allows estimation of the expected values for installing sand control or not and improves judgments of the true cost of obtaining the DST information.

Wellbore Stability Analysis Based on the Combination of Geomechanical and Petrophysical Studies

The COVID-19 virus exhibits pneumonia-like symptoms, including fever, cough, and shortness of breath, and may be fatal. Many COVID-19 contraction experiments require comprehensive clinical procedures at medical facilities. Clinical studies help to make a correct diagnosis of COVID-19, where the disease has already spread to the organs in most cases. Prompt and early diagnosis is indispensable for providing patients with the possibility of early clinical diagnosis and slowing down the disease spread. Therefore, clinical investigations in patients with COVID-19 have revealed distinct patterns of breathing relative to other diseases such as flu and cold, which are worth investigating. Current supervised Machine Learning (ML) based techniques mostly investigate clinical reports such as X-Rays and Computerized Tomography (CT) for disease detection. This strategy relies on a larger clinical dataset and does not focus on early symptom identification. Towards this end, an innovative hybrid unsupervised ML technique is introduced to uncover the probability of COVID-19 occurrence based on the breathing patterns and commonly reported symptoms, fever, and cough. Specifically, various metrics, including body temperature, breathing and cough patterns, and physical activity, were considered in this study. Finally, a lightweight ML algorithm based on the K-Means and Isolation Forest technique was implemented on relatively small data including 40 individuals. The proposed technique shows an outlier detection with an accuracy of 89%, on average.

Machine learning-assisted design guidelines and performance prediction of CMOS-compatible metal oxide-based resistive switching memory devices

The Attaka Field is a giant oil field in offshore East Kalimantan. Reservoirs in this field are multi-layered sandstones eposited in deltaic to shallow marine environments during Middle to Late Miocene time. Within this field, high-resistive as well as low-resistive hydrocarbon bearing sandstone layers are present. Low resistivities occur in both thick-and thin-bedded sandstones, implying that shoulder-and dipping-bed effects are not the only reasons for the low resistivity readings. High irreducible water saturations in micro pore spaces and bound water associated with high clay content also lower the resistivity of pay sands. In the thin bedded shaly reservoirs, shoulder-bed effects have complicate low-resistivity problems. The conventional logs, such as the dual and phasor induction, cannot resolve true resistivities in sand layers less than 5 feet thick. To derive resistivities that are closer to the true formation resistivities, Unocal employs an in-house deconvolution program for dual induction and other older logs, and an enhancement program for phasor induction logs. Unocal has tested a magnetic pulsed induction log which shows higher resolution than the older induction logs. In addition, utilization of high sampling rate lithodensity and neutron logs results in more accurate porosity values in thin bedded reservoirs than can be achieved with older porosity logs. Determination of fluid contents of low-resistive sandstone layers is based on qualitative log evaluations, correlations, Repeat Formation Tester (RFT) samples, and Drill Stem Tests (DST). Formation evaluations also incorporate core analysis data. Using new logging technology and integrated formation evaluations, Unocal has mapped a significant increase in reserves from the low-resistive sandstone layers. These reservoirs are among the primary targets for the current development plan.

Software designers are coming up with tools having machine learning (ML) capabilities integrated into them. Even naive users can build ML-based models for their respective problem domains using these tools. However, a majority of the advanced ML models behave like a black box in the sense that their behavior is not easily interpretable by humans. This lack of human interpretability acts as a hurdle toward acceptance and subsequent deployment of ML-based solutions. This research work is based on an intuitive idea that in an ideal ML-based solution, the provided dataset, the ML model learned using this dataset, and human domain experts must agree in terms of their perception regarding the problem domain. The objective of this work is to propose a framework that first listens to the dataset, the ML model, and human domain experts individually, and then measure the degree of agreement between them. Listening to the dataset refers identifying important characteristics by computing information gain using entropy and Gini index. Listening to the ML model refers interpreting its decision-making behavior by computing variable importance measures. Listening to human experts refers taking feedback in terms of important features as per their perception regarding the problem domain. The task of measuring agreement between dataset, model, and human experts has been modeled as a “two-judges n-participants” rank correlation problem. The proposed approach has been evaluated on a novel problem domain of predicting joining behavior of freshmen students. A positive degree of agreement was observed between the dataset, the ML model, and human experts in terms of their perception regarding the problem domain. The agreement between the provided dataset, the learned model, and human domain experts helps verify learning acquired by the ML model against the prevailing domain knowledge. This work has the potential to form the basis for developing formal quantitative metrics for evaluating ML models in terms of reliable learning and capability to facilitate trust of human users.

Due to the latest advancements in monitoring technologies, interest in the possibility of early-detection of quality issues in components has grown considerably in the manufacturing industry. However, implementation of such techniques has been limited outside of the research environment due to the more demanding scenarios posed by production environments. This paper proposes a method of assessing the health of a machining process and the machine tool itself by applying a range of machine learning (ML) techniques to sensor data. The aim of this work is not to provide complete diagnosis of a condition, but to provide a rapid indication that the machine tool or process has changed beyond acceptable limits; making for a more realistic solution for production environments. Prior research by the authors found good visibility of simulated failure modes in a number of machining operations and machine tool fingerprint routines, through the defined sensor suite. The current research set out to utilise this system, and streamline the test procedure to obtain a large dataset to test ML techniques upon. Various supervised and unsupervised ML techniques were implemented using a range of features extracted from the raw sensor signals, principal component analysis and continuous wavelet transform. The latter were classified using convolutional neural networks (CNN); both custom-made networks, and pre-trained networks through transfer learning. The detection and classification accuracies of the simulated failure modes across all classical ML and CNN techniques tested were promising, with all approaching 100% under certain conditions.

Lithology identification holds a pivotal role in the characterization of subsurface formations. In recent years, the advent of big data and the need for more precise lithology identification have spurred the growing adoption of machine learning algorithms. This paper's primary objective is to leverage unsupervised machine learning techniques for the identification and subsequent removal of anomalies inherent in complex datasets. The Isolation Forest model is the cornerstone of our approach for anomaly detection and elimination. To initiate this process, we employed the K-means algorithm to create clusters, followed by a evaluation using silhouette coefficients. Subsequently, we selected input data for each cluster and conducted exploratory data analysis both prior to and after the removal of outliers. The histograms portraying average anomaly scores for each cluster were thoughtfully presented. For real-time anomaly detection, we harnessed the power of the Isolation Forest model, which facilitated the creation of an Isolation Forest anomalies map by plotting neutron-porosity against bulk density. Our rigorous data analysis employed various statistical techniques, including data statistics, histograms, and cross plots between neutron-porosity and bulk density. This multifaceted approach successfully eradicated anomalies from the dataset, a fact vividly illustrated by the histograms, where anomalies were discernible through their negative scores. The Isolation Forest map conclusively demonstrated the effective removal of outliers from the dataset, underscoring the model's proficiency in identifying and mitigating these anomalies based on their negative scores. The Isolation Forest model has thus exhibited remarkable efficacy in the identification and elimination of data anomalies. Its versatility makes it a valuable asset for the detection and removal of outliers, deviations, or noise from datasets, rendering it particularly well-suited for anomaly detection and outlier mitigation in various analytical scenarios. Notably, the combination of the K-means and Isolation Forest algorithms emerges as a potent and advantageous approach, especially when dealing with extensive datasets and conducting comprehensive analyses.

The Norwegian Central Graben contains a distinctive three-tiered sub-surface pressure system. This pressure system has an important effect on distribution of oil and gas fields, generation and migration of hydrocarbons, reservoir qualities and drilling procedures. A number of mechanisms have been proposed for the development of overpressure. In this paper, five are briefly discussed: thermal expansion of water, sedimentary compaction, rate of sediment burial, volumetric increases due to generation of hydrocarbons, and column heights of major oil and gas fields. None of the factors are mutually exclusive and all contribute to the development of overpressure in the Norwegian Central Graben. Utilizing drill-stem tests (DSTs), repeat formation tests (RFTs) and sonic and density logs, the horizontal and vertical boundaries of the three-tiered pressure system of the Norwegian Central Graben have been mapped. A shallow hydropressure system exists down to 1000–2000 m, underlain by a transition zone where pressures abruptly increase to the fracture level where overburden and fluid pressure are in balance. Under the transition zone, a geopressure zone exists to 3000–4000 m, within which pressures increase with depth at a hydrostatic rate. In the centre of the graben, this compartment is underlain by a second transition zone and a deep, high temperature, high pressure zone. The location and depth of the boundaries of the two upper pressure compartments and transition zones are more dependent on temperatures than lithologies or tectonic movements. Within the deep, high pressure zone, variations in pressure gradients indicate the presence of a number of smaller pressure cells, probably separated by faults. The identification and prediction of overpressured zones can have a significant effect on exploration programmes. From the drilling standpoint, prediction of sharp increases in pressure is a major safety factor, reducing the danger of well blowouts and allowing the operator to be adequately prepared for a number of drilling hazards. In potential reservoirs, overpressure reduces net overburden, thereby reducing porosity loss due to compaction. Sub-surface pressure variations have a significant effect upon hydrocarbon migration as oil and gas will attempt to move from high to low pressure compartments. Defining pressure variations from well to well can give indications of sealing and non-sealing faults. The pressure configuration has a significant effect upon the distribution of discovered hydrocarbons in the Central Graben. Most recoverable hydrocarbons are found in the first sealed permeable bed in the middle pressure compartment, overlying or adjacent to the transition zone above the deep, high pressure compartment. This includes the Cretaceous to Danian chalk fields in the centre of the graben and the Jurassic sand-reservoired fields on the northeast flank. Exploration in the deep, high pressure compartment has resulted in only limited success due mainly to poorer potential reservoir conditions, more difficult definition of traps and increased drilling expenses and risk.

The success of Machine Learning (ML) approaches as promising solutions has encouraged their widespread implementation across different fields. Owing to the high accident rate, the construction industry embraced ML in the risk assessment procedure. What if the machine produces knowledge of the relationship between the risk features and accident outcomes contained in the safety dataset? What if machines can explain an accident dataset without human intervention? Unsupervised ML techniques offer several advantages over supervised approaches, including their explainability to analyze and understand complex datasets. This chapter demonstrates the practical implementation of the unsupervised learning method, clustering, and dimensionality reduction to explain the similarities, differences, variances, and patterns that exist between the feature spaces of an occupational safety risk dataset. Principal component analysis (PCA) and K-means clustering with silhouette analysis were selected as two unsupervised ML approaches to demonstrate their implementation in enhancing data-centric decision-making during the construction risk assessment procedure.

Customer relationship management (CRM) is the tool to enhance customer relationship in any business. Due to the exponential growth of data volume, in any field, it is significant to develop new techniques to discover the customer knowledge, automation of the system and moreover customer satisfaction to win customer lifetime value. CRM with machine learning could bring a catalytic change in business. Several supervised and unsupervised machine learning techniques are utilized to improve the customer experience and profitability of business. This paper reviews the available literature on the CRM with machine learning techniques for customer identification, customer attraction, and customer retention and customer development. This study reveals that supervised learning techniques are 48.48% utilized, unsupervised learning techniques are utilized 15.15%, and 9.09% utilized other techniques in CRM. Paradigm is also shifted toward the deep learning from machine learning as 28.28% text has been reported to deep learning. Decision tree-based algorithm and support vector machine algorithms are most utilized algorithm of supervised learning. E-commerce and telecommunication sectors are the most important areas identified with the exponential growth of the users and hence need a suitable machine learning techniques for customer satisfaction and business profitability.