... In addition, redundant data samples and properties are dropped. For instance, in this specific dataset the injection data of water and gas was constant across all realizations and was thus eliminated. In fact, the ML model learns to predict production data across all wells which have different...

... in predictions affected by various factors. This paper presents a step-by-step workflow of applying a ML approach to develop a heterogeneous permeability prediction model from the CT images of core samples. In this work, over ten thousand 3-D sub-image were randomly extracted from the CT images of two...

... by tuning the hyperparameters. This feature of gradient boosting renders it one of the most powerful algorithms for predicting output parameters. Extreme Gradient Boosting Algorithm Decision Tree Algorithm The classic decision tree model is the simplest form of the tree-based algorithm. During...

... to classical Arps models ( Arps, 1945 ), many models have been developed to forecast well performance, and recent models have been dedicated to predicting the production of horizontal wells in unconventional reservoirs ( Valko and Lee 2010 , Agarwal et al. 1999 , Fetkovich 1980 , Ali and Sheng 2015...

... and at the same time, reliable alternative to the conventional numerical simulators. In this work, we have developed a DL approach to effectively predict the dissolution and precipitation of various important minerals, including Anorthite, Kaolinite, and Calcite during CO 2 injection into deep saline aquifers...

... and correspondingly adjust the low-resolution profile to ensure mass-balance. Table Table 2 provides the performance difference with each type of data augmentation. Abstract Complete physics-based numerical simulations currently provide the most accurate approach for predicting fluid flow behavior...

... was obtained with predictive capabilities to forecast the field's behavior. Then, we combined AI-Physics history training with blind test prediction calculation of remaining oil maps. Finally, forecast scenario definitions based on the remaining oil map were created by the AI-Physic model. The proposed novel...

... an accurate prediction. The obtained ANN model was compared with the vapor–liquid equilibrium (VLE) calculation and was used as a surrogate model for accelerating the flash calculation. The comparison revealed that the ANN model reduced the computational time of the VLE calculation by as high as 250 times...

... with additional samples or reduced by removing redundant samples (simulation runs). Consequently, the hybrid neural approach also provides a clear picture of which simulation runs (or samples) are more conducive to optimal recovery predictions for an effective strategy to sample the high dimensional WAG parameter...

... (measured log vs predicted log) show different elastic properties, each crossplot shows the CC (correlation coefficient) and NRMSE (Normalized Root Mean Square Error). In the lower-right, different elastic crossplots were generated, the black points are the measured data and in grey are the modelled data...

... of the confinement of plumes at each potential storage site ( Cook et al., 2014 ; Keating et al., 2016 ; Pawar et al., 2014 ; Shabani and Vilcáez, 2018 ). The accurate prediction of the flow, geochemical, and geomechanical responses of the formation is essential for the management of GCS in long-term operations...

... Abstract This paper discusses the adoption of Machine Learning (ML) approach to identify new Behind Casing Opportunities (BCO) in two brown fields (B and S) offshore East Malaysia. A multi-stage field-based ML models were developed based on selected wells and consequently used to predict...

... reservoir geological parameters to reservoir decision parameters by coupling it with f l ; (2) directly optimizing the reservoir decision parameters based on coupling the existing optimizers such as Adam with f l . The forward model f l performs accurate and stable predictions of evolving temperature...

..., while it often suffers from nonphysical solutions and unexplainable models. The presented method holds the properties of explainable regression models while providing powerful predictability capabilities within material balance constraints. By no means does it try to replace the reservoir simulation...

... (LS-SVR). Both proxies have the capability of predicting well outputs. The sequential quadratic programming (SQP) method is used to perform nonlinearly constrained production optimization. The objective function considered here is the net present value (NPV), and the nonlinear state constraints...

... upscaled by averaging grid block properties. The AI simulator will predicts the behavior of reservoir with the art of enriching the AI engine with the overall physical behavior from dynamic modeling. To get this overall behavior only an upscaled dynamic model is required which can be prepared quite...

... of the surrogate models. The developed workflow showed superior performance by giving AAPE less than 5% and R 2 more than 0.95 between ground truth and predictions of the state variables. The proposed Deep Learning framework provides an innovative approach to track CO 2 plume in a fractured carbonate reservoir...

.... The even bigger challenge is to make these calculations fast, especially given the expectation that a typical saline aquifer model will be many times larger areally than used for oil and gas production. Various methodologies have been published to predict the phase behavior and fluid properties of CO 2...

... to guide new wells location and trajectory. Actually, two proof of concept case histories have demonstrated the reliability of this approach. The newly drilled wells, with optimized paths according to these prediction-maps, have intercepted the desired karst intervals as per the subsequent image log...

... as intrinsic noise. Consequently, using a machine learning-based method that can resolve both the temporal and spatial non-linear variations has advantages over a pure engineering model. The approach followed provides a long short-term memory (LSTM) network-based methodology to predict BHP and temperature...