... interpretation data mining dataset aibk 0 node cement and bond evaluation algorithm software metric cement integrity assessment casing and cementing artificial intelligence probability specialist supervised machine learning applied application operation prediction Introduction The Oil...

... is built on a combination of physics-based data augmentation and unsupervised data learning method. The model is applied on multiple field bit runs. Results show that by implementing the unsupervised data learnings only, the prediction accuracy is improved by 30% compared to the baseline ML model...

... these approaches. It does so by employing appropriate data normalization methods, transfer learning, and the inclusion of physics-based features. Using historical offset well data, this paper presents trained and tested machine learning models capable of predicting the stick-slip index (SSI) using sequences...

... geologist geology artificial intelligence neural network dht machine learning circulation dht prediction geothermal well operation scenario drilling fluids and materials prediction dataset conductivity deep learning drilling fluid management & disposal validation...

... Different ML algorithms were utilized in order to achieve better prediction results. Since the initial results only provided 70% accuracy, there was a necessity for collecting new parameters. Figure 2 shows the workflow of feature collection and selection. Abstract Wells with compromised...

... As new logs from the currently drilled well become available, the PSI input parameters are updated; in particular, an important update source is the pore pressure prediction obtained by running the Pressim2.0 software prior to drilling and updated with resistivity and sonic log readings. Using...

... prediction ( Bin Othman et al., 2022 ; Meor Hashim et al., 2021 ) and early kick detection ( Wang et al., 2023 ) based on recurrent neural networks (RNNs). Training a classification model to distinguish between intervals of real-time signals that are likely to lead to incidents and those...

... of these trajectories. Abstract Many models in the industry use the notion of an effective length to make predictions about the bottom hole assembly (BHA), whether for its directional behavior or its lateral vibration tendency. A given criteria is chosen to cut the drill string at a certain distance from the bit...

... classification scheme. This in turn opened for more opinion-based and subjective interpretations. Abstract The Assisted Cement Log Machine Learning (ML) tool – or Annular barrier AI tool - developed by Equinor - is being used to interpret cement logs by predicting a predefined set of annular condition...

... is used to determine the optimal combination of drilling parameters that yield above-average ROP performance. Robinson et al. (2022) presented a case for a "global" ROP prediction model which does not require retraining models for every new application. A deep neural network ML model was trained...

... the drill bit to penetrate the rock during drilling a well. Correctly predicting mud motor failure and the remaining useful life of the components are essential for obtaining drilling efficiency, avoiding costly operational expenses, and achieving timely maintenance. The remaining useful life indicator...

..., such as the classical broomstick plot or alarm triggering. A hybrid T & D modelling framework was designed to predict the hook load and surface torque accordingly. This framework combines the mathematical capability of a stiff-string model using a finite element method and the experience acquired from obtaining...

... and execute safe, optimum drilling operations. In this context, "accurate" is commonly understood as 0.1 pound per gallon (ppg) delta, or less, between measured and predicted equivalent circulating density (ECD). Extended-reach-drilling (ERD) wells have been gaining recent adoption due to advancements...

... artificial intelligence models to make predictions and analyze trends and perform patten recognition. Models built using artificial intelligence offer several advantages over traditional models. They have the ability to learn on the go and adapt to changing situations which enables the models...

...-208743-m prediction rop performance bit size optimization model different drilling environment threshold optimization process rop implementation provide consistent improvement Introduction After decades of drilling experience, the Oil and Gas industry has managed to optimize processes...

...Introduction The ability to predict WOB and RPM ranges with associated stick/slip classes and their probabilities will be more valuable to a driller than the ability to predict Slip/slip given a set of input parameters. These predictions will provide the drilling operator with an understanding...

... is proportional to the exerted frictional work by the rotating tool joint at the same point ( Mittal, Samuel, 2019 ). Design considerations for casing wear factor prediction The present paper aims to present the newly designed and built casing wear testing facility by repurposing and automating an lathe...

... drilling (MWD) data as input to predict reamer vibration. Wipeout risk levels are calculated and sent to the operators for real-time decision making to reduce the possibility of reamer wipeout. Large volumes of reamer field data, including field recorded vibration and reamer dull conditions were used...

..., the predicted pressure cannot be considered accurate enough in every condition ( Cayeux 2012 ). Frequently, the simulated and measured pressure do not match, thus an automated detection of abnormal pressure based on this pressure difference cannot be performed with this approach. Abstract Abnormal surface...

... learning ROP model. Lagged features have been used in other industries for time-based predictions. Simply put, lagged features are the most recent values of the inputs to the value being predicted, or derivations of the output itself. Another way to look at it is that in time-based series what is about...