...Ensemble Machine Learning model An ensemble model has proved to be very effective and extremely versatile to address a variety of machine learning problems for building predictive model. The ensemble machine learning method is famous on improving model accuracy originally by reducing...

... fracturing u-net architecture artificial intelligence upstream oil & gas training dataset recognition neural network ball pumpdown seat operation prediction ball seat event recognition algorithm dataset slurry rate machine learning cnn model wellhead pressure computational resource deep...

... deciphered, combined and organized into a structured database. Imputation techniques were used to fill the gaps of missing data. With the Augmented AI approach, the median accuracies of IP and EUR predictions for new drills is around 90%, which often outperforms industry-standard type curving methods...

... Abstract In today's data-driven economy, operators that integrate vast stores of fundamental reservoir and production data with the high-performance predictive analytics solutions can emerge as winners in the contest of maximizing estimated ultimate recovery (EUR). The scope of this study...

... accurately predict the transient between annular and slug flow regimes in horizontal wells. As more wells are produced in shale gas reservoirs, a great amount of information from production control and monitoring becomes available which can be used to build a data-based smart model for real time diagnostics...

... on kriging and Markov Chain Monte Carlo (MCMC) sampling algorithms were utilized to integrate data sets and predict geological properties at production well locations. The spatial Gaussian Process regression modeling was conducted to investigate the primary controls on production. Results suggest that the 6...

... significantly change the reservoir behavior prediction. Moreover, results suggest that at initial stage high compressibility value will contribute to production significantly. However, at the same time, this high compressibility in shale matrix, will lead to pore throat closure, thus loss of connectivity...

..., and consequently the fracture modeling results. To accurately predict the hydraulic fracture propagation and its geometry (height, length, and width) from petrophysical logs in this environment, these mechanical property differences must be considered. This is achieved by predicting, from petrophysical logs...

... permeametry, mercury porosimetry, helium pycnometry and scanning electron microscopy (SEM) images performed on 53 samples of the Montney Formation to understand the complicated pore network structure of the rock and study the predictive power of a permeability prediction model. Pulse-Decay permeability...

... for prediction. Some common data mining tools are neural networks (NN), genetic algorithms (GA), and self-organizing maps (SOM). Our method uses NN for predicting the future performance of a shale gas well based on historical production data of the previous year. The decline in production is captured during...

... on reservoir development and profitability. risk management Upstream Oil & Gas risk and uncertainty assessment Artificial Intelligence risk assessment drilling operation management actual total well cost overestimation Engineer well cost prediction drilling campaign historical data...

... years of production history. For wells having less than 2 years of production history, the YM-SEPD Method will also be able to yield a reasonable prediction by coupling with Duong’s Decline Method production control shale gas production forecasting production monitoring complex reservoir...

... Abstract In this paper, moment of momentum theorem is introduced and used to model the ratio of the rotary speed of cone to the bit. The ratio model has been integrated in a ROP simulator and the simulator can predict the ratio of both tricone and monocone bits. A series of tests measuring...

... reservoirs, and compare the apparent permeability and slippage factor predictions for all the models we have compiled. For determining the impact of the model predictions on well performance, we incorporate the apparent permeability predictions of each model into a numerical simulator, and compare rate-time...

... it is difficult to determine which model to use in predicting the production rate. Based on a broad literature survey, three shale models were selected to evaluate against field data. The three models are the Duong model, the stretched exponential production decline model and the fracture flow model. The field...

... prices. Starting with the current spot price, the model will generate a random but equiprobable prediction of future prices. The model can be used as input into a Monte-Carlo simulation. Alternately, the model can be run multiple times in order to generate "high", "low", and "expected" price predictions...

... Abstract Accurate predictions of heavy oil and bitumen viscosity as a function of temperature, pressure, and composition are required for the design of thermal and solvent based recovery methods. In this case study, the applicability of the recently developed Expanded Fluid viscosity...

... asphaltene precipitation prediction mole fraction experimental data precipitate multiphase molecular weight asphaltene deposition composition badamchizadeh fraction equation mole fraction nc7 Introduction The worldwide original oil in place of heavy oil and bitumen is estimated...

... . SeTES uses as much (or as little) information as the user desires in order to make recommendations about well location, orientation and treatment schedule. If no field data is selected but parameters of a theoretical reservoir are entered, then SeTES automatically switches to a predictive mode...

... and gas velocities. Tests were conducted at oil viscosities of 0.587, 0.378, 0.257 and 0.181 Pa-s. Superficial liquid and gas velocities varied from 0.1 to 0.8 m/s and 0.1 to 3.5 m/s, respectively. The experimental measurements were compared with the existing slug liquid holdup model predictions...