NobleAI’s Post

Very excited to see NobleAI in this article. It was such a pleasure to work with Colter Cookson on this rather complex topic. If you want to good understanding of the future of AI in material development and all things chemical, please read on!

Machine Learning (ML) When data are brought together for a given model, whether, Machine Learning (ML) will be able to figure out the areas, where field measurements are required? Whether ML could accommodate the application of physical laws to field data, which would possibly reveal additional information about 'unmeasured' or 'difficult to measure' field properties? Whether ML could offer insight to the system being modelled? At least, ML could act as a Parsimonious model for any given physical system, which are based on the simplest conceptual mechanisms and employ fewest parameters, while also providing an acceptable representation of a given physical system by providing the basic insights to the system functioning and critical processes? Whether the forecasts from ML could test hypotheses about system responses and allow quantitative comparisons of alternative proposed scenarios? How exactly ML is expected to improve the model performance, given the fact that the complexity of petroleum reservoir systems and the uneven spread, poor quality or even absence of observed data present considerable difficulties for oil/gas drainage modelling? Suresh Kumar Govindarajan https://lnkd.in/d6rtS6Ue https://lnkd.in/d_miY7ZU

The Impact of AI on Petroleum Engineering

* Enhanced machine learning-ensemble method for estimation of oil formation volume factor at reservoir conditions: https://lnkd.in/dkmTJj7g Abstract: Since the oil formation volume factor (Bo) is crucial for various calculations in petroleum engineering, such as estimating original oil in place, fluid flow in the porous reservoir medium, and production from wells, this parameter is predicted using conventional methods including experimental tests, correlations, Equations of State, and artificial intelligence models. As a substitute to conventional black oil methods, the compositional oil method has been recently used for accurately predicting the oil formation volume factor. Although oil composition is essential for estimating this parameter, it is time-consuming and cost-intensive to obtain through laboratory analysis. Therefore, the input parameter of dissolved gas in oil has been used as a representative of the amount of light components in oil, which is an effective factor in determining oil volume changes, along with other parameters, including pressure, API gravity, and reservoir temperature. This study created machine learning models utilizing Gradient Boosting Decision Tree (GBDT) techniques, which also incorporated Extreme Gradient Boosting (XGBoost), GradientBoosting, and CatBoost. A comparison of the results with recent correlations and machine learning methods adopting a compositional approach by implementing tree-based bagging methods: Extra Trees (ETs), Random Forest (RF), and Decision Trees (DTs), is then performed. Statistical and graphical indicators demonstrate that the XGBoost model outperforms the other models in estimating the Bo parameter across the reservoir pressure region (above and below bubble point pressure); the new method has significantly improved the accuracy of the compositional method, as the average absolute relative deviation is now only 0.2598%, which is four times lower than the previous (compositional approach) error rate. The findings of this study can be used for precise prediction of the volumetric properties of hydrocarbon reservoir fluids without the need for conducting routine laboratory analyses by only employing wellhead data.

Application of Machine Learning (PIML) in Petroleum Reservoir Engineering Applications With reference to the availability of Reservoir Rock Properties, we use to have a relatively lesser data for Porosity (despite having normal distribution of data); very less data for Permeability (obviously due to its log-normal distribution @ field-scale); and almost nil data for Rock Compressibility (becomes critical in layered formations), with reference to the availability of the relatively larger Reservoir Fluid Properties. Further, @ field-scale, we can never measure the fundamental multi-phase fluid flow data properties @ sub-pore scale, associated with capillary forces (capillary pressure: IFT) and wettability (contact angle); and, we can only get “equivalent” values from relatively smaller laboratory-scale investigations using experimental techniques (lab data estimated, not at the required scale of Darcy or continuum), in the absence of any reservoir physical and chemical heterogeneities. 1.  So, with huge data noise; with heterogeneous data noise; with significant outliers; and with drastic distribution shift in data; along with data sparsity associated with several fluid regimes of the reservoir; and with a randomness in data partitioning; how will we be able to quantity the sources of uncertainty associated with Reservoir Rock Properties (data side)? 2.  And, in the case of fundamental multi-phase Reservoir Fluid Flow parameters (IFT, contact angle), we do not even have the real field-data, while, we just manage these parameters using lab-scale investigations. If so, then, how could we justify ‘Data Quality’ in Reservoir Engineering applications? 3.  While dealing with the strongest prior, i.e., Partial Differential Equations in the case of Reservoir Engineering applications, (unlike dealing with the relatively weaker priors like Ordinary PDEs, or, stochastic DEs, or, symmetry constraints, or, intuitive physics constraints); while, given the poor justification towards Data Quantity and Data Quality, how will we be able to deduce the Reservoir "Architecture" from such data sets? 4.  With poor architecture, how would it remain feasible to deduce ‘Improved Loss Functions’ from its translation from data-preprocessing, deep neural network, data-processing and automatic differentiation? What exactly is happening during 'Automatic Differentiation' and 'Physics Information' modules (from DNN output)? 5.  How will we be able to deduce "individual weighting functions" associated with PDEs, initial conditions, boundary conditions and data? 6.  With compromised loss function estimation, how would be able to go ahead with reasonable ‘optimization’, before, we start ‘inferring’ the generated results towards forecasting – given, either the neural simulation or an inverse problem? Suresh Kumar Govindarajan Professor (HAG)   IIT-Madras https://lnkd.in/d6rtS6Ue https://lnkd.in/d_miY7ZU 17-August-2024

AI plays a significant role in petroleum geology by helping geoscientists and engineers analyze large volumes of data to make informed decisions for oil and gas exploration and production. Here are some ways AI is utilized in petroleum geology:Seismic Interpretation: AI algorithms can process seismic data to identify potential hydrocarbon reservoirs beneath the Earth's surface by detecting patterns and anomalies.Reservoir Characterization: AI can analyze well logs, core data, and reservoir simulations to understand the properties of a reservoir, such as porosity, permeability, and fluid compositions.Production Optimization: AI algorithms can predict production decline curves, optimize well placement, and identify opportunities for enhanced oil recovery techniques.Drilling Optimization: AI can assist in real-time drilling operations by predicting potential drilling hazards, optimizing drilling parameters, and improving wellbore placement.Predictive Maintenance: AI can help predict equipment failures in oil and gas facilities, enabling proactive maintenance to minimize downtime and maximize production efficiency.Overall, AI in petroleum geology can enhance decision-making, reduce exploration risks, improve operational efficiency, and ultimately lead to increased oil and gas production.

The petroleum industry has long had to deal with "Big Data," acquired from various well logs and geologic methods (seismic, gravity, magnetic), but using this information wasn't always an easy task. How can we use this data to make better predictions in oil and gas settings? A paper from ScienceDirect, named Petroleum Research, dives deeper into the applications of machine learning and AI in the oil and gas industry. This is a long paper that covers many topics, such as algorithms, exploration, production engineering, and examples of such analysis. As a startling fact, it is said that 1 in 7 wells in the past (before modern technological analysis) hit oil using traditional well location methods, but advancements in data analysis tools have reduced that number to merely 1 in 3 in situations. If you're even wondering what tools in machine learning are used for what predictions, I would like to show this chart below straight from the article. It was insightful for me as I hope it is for you too. | Subsurface Geology |: Find characteristics of reservoir, accelerating well log collection, rock typing tools (use interpolation, gradient boosting) | Drilling |: Detect rock form and failure (use DNN and ensemble learning) | Reservoir Engineering |: Speed up traditional reservoir simulation tools using DNN | Production Optimization |: Determine well efficacy using gradient boosting and feature selection from professionals #PGE301 #machinelearning #oilandgas https://lnkd.in/gKnuCDdn

I am delighted to announce our latest publication focusing on the application of AI in estimating reservoir parameters in the International Journal of Petroleum Exploration and Production Technology, Springer. This work tackles the challenges arising from the lack of essential core data. We provide validation and evaluation of our approach, documenting its reliability in reducing estimation errors. Moreover, this study demonstrates the significant potential of AI-based approach in solving reservoir characterization and evaluation associated problems. A sincere thanks to the esteemed professors, Dr. Golnaz Jozanikohan and Dr. Ali Moradzadeh for their kind support. To access the publication, use the link below. #petroleumengineering #reservoircharacterization #AI

* Machine learning-based cement integrity evaluation with a through-tubing logging experimental setup: https://lnkd.in/dCQQ7EBB Highlights: • Experimental setup mimicking a multi-string oil well section was designed. • Proprietary sonic logging tool performing through-tubing measurements. • Logging runs performed for sixteen different wellbore conditions. • Machine learning classification framework developed to assess cement integrity. • SVM model achieved accuracy of 99% and prediction time inferior to 1 ms. Abstract: Assessing the integrity of the cement layer and the quality of its bond to the casing and formation is paramount to ensure that the wellbore is hydraulically isolated from the surrounding environment before permanently sealing the well. Such inspection is part of Plugging & Abandonment (P&A) operations, and it is usually achieved through well logging tools, which provide a vast amount of data that a skilled specialist interprets. The process is human-dependent, error-prone, and time-consuming. There has been an increasing interest in solutions that allow an automatic interpretation of the logging data since the recent panorama of the oil and gas industry indicates a growing P&A demand. Such solutions aim at reducing the dependence on human knowledge and consequently increasing the reliability and accuracy of the cement integrity evaluation. Therefore, this work presents an experimental setup capable of emulating defective cement layer configurations of real-world oil wells in single or multi-string arrangements. Such flexibility enables acquiring logging data for multiple well conditions and building a rich logging database to develop a supervised learning framework and define the most suitable model for performing the cement integrity evaluation. Several logging runs were performed, producing a database with 130 samples, including varied tubing eccentricity levels and cement layer conditions. A complete analysis of the data both in the time domain as well as in the frequency–wavenumber domain was performed, highlighting the complexity of the interpretation task. A resampling-based workflow was employed to evaluate machine learning models of different families. The models were tested under three scenarios, and accuracy and computational complexity metrics were computed to compare their performance. The results showed that shallow learning models can perform satisfactorily well even with less data available for training. The support vector machine stood out, achieving a mean accuracy score higher than 0.99 while being able to predict the cement sheath’s condition in less than 1 ms. This paper contributes to the research on the cement integrity evaluation by presenting a study that combines an experimental setup mimicking several oil well conditions and the employment of machine learning as a diagnostic tool, which has no precedents.

⚡ Modeling crude oil pyrolysis process using advanced white-box and black-box machine learning techniques | Nature Discover more at "voltsbase.com/energy" #Energy #Voltsbase