... co 2 iteration corrosion prediction Introduction In the energy industry, an enormous infrastructure of pipelines underpins the wide spectrum of operations involved with the production, transport and processing/productization of hydrocarbon resources as well as emerging energy sources...

... by setting α = 0. Abstract Previous implementation of distributed Gauss-Newton (DGN) optimization algorithm runs multiple optimization threads in parallel, employing a synchronous running mode (S-DGN). As a result, it waits for all simulations submitted in each iteration to complete, which may...

... represent minimum and maximum values of the NPV, respectively, and are obtained from the training set. Since during the iterative-sampling-refinement optimization process, the training set changes, J min and J max may change from iteration to iteration. Optimization Methods In this section, we...

... computed is presented. All results were obtained during the validation tests of the algorithm. evolutionary algorithm upstream oil & gas artificial intelligence bit position particle presented machine learning drilling operation well plan footage iteration drilling window...

... learning artificial intelligence reservoir simulation constraint iteration well location drilling operation trajectory well trajectory automated approach applied oil recovery optimized location Introduction Identifying optimal well configurations is crucial for maximizing recovery from...

... multiple local optima of highly nonlinear optimization problems. However, its performance has neither been validated by realistic applications nor compared to other DFO methods. We have integrated DQN into a versatile field-development optimization platform designed specifically for iterative workflows...

... are calculated with iterations with a numerical method to match the GOR and WOR considering the wellbore pressure and temperature effect for each section. Once all the sections are calculated, the final results are combined to achieve the global gas/oil and oil/water interfaces and the pressure and temperature...

...) techniques to reduce the number of simulations for optimizing long-term well control strategy while preserving the quality of the solution. We proposed a methodology, denoted as IDLHC–ML, which combines many ML techniques with iterative discrete Latin hypercube (IDLHC) – a gradient-free optimization...

.... It is an iterative process, which yields the correct initial gas in place, whereas the use of real time does not. The algorithm is validated with a two gas well synthetic example, and successfully applied to a field case with four gas producers. Novel/Additive Information The paper extends the application...

... iterations, a clear straight line passing through the origin that is tangent to the semi-log derivative of the pressure vs. semi-log derivative curve was observed. Closure pressure can be estimated at the deviation point of the semi-log derivative of pressure versus G-function from the straight line through...

... curve pore size distribution relative permeability curve Upstream Oil & Gas iteration capillary pressure curve reservoir production mode Introduction and Review of Past Work Relative permeability curves are utilized in reservoir simulation to determine relative flow rates when...

... and enables robust business decision makings under uncertainty. machine learning Artificial Intelligence history matching reservoir simulation hydraulic fracturing Upstream Oil & Gas optimization problem objective function model parameter fracture height simulation result iterative...

...Overview of Probabilistic History Matching Methods The foundation of the Ensemble Smoother follows Bayes' rule and thus the final finite n samples obtained from iterative process describes the posterior distribution. Unlike rejection sampling, analytical formula is used to update the model...

... overpull pipe weight drag model detection torque hook load trend society of petroleum engineers hook load value iteration hook load probability operator Introduction Traditional electronic drilling recorders (EDR’s) are third party systems that collect rig sensor data and make...

...-section image fast bayesian conventional approach recovery recovery performance image reconstruction omp optimization problem pursuit algorithm cosamp cs greedy pursuit algorithm greedy pursuit algorithm society of petroleum engineers Exhibition iteration image recovery algorithm...

... which are used in Bayesian optimization to find history match solutions in an iterative manner. The uncertainty in history match solutions is quantified via MCMC sampling on trained forward GP model. In order to check for independence and convergence of the collected MCMC samples, auto-correlation plots...

... is further improved by reusing simulation results of preceding iterations. Starting from an initial ensemble of models, new search points for each realization are generated with a modified Gauss-Newton trust region method using the sensitivity matrix estimated with SVR. The SVR proxies are updated when...

... Choose N i . Definition of the number of iterations (N i ) Initial Ensemble. To generate the initial ensemble, it is necessary to define the amount of models (N e ) that compose the ensemble. Uncertainties and its parameterization are delineated to generate pdfs. To sample the values...

... The method used to update uncertainty differs depending on the type of attribute. For discrete attributes, we combine the current pdf with the frequency of occurrence from the best-matched models to generate a new distribution, which is used in the next iteration. For attributes with spatial...

... superposition Reynolds uncertainty quantification covariance matrix iteration history matching Gaussian distribution toy problem ensemble Kalman filter normalized objective function map point realization posterior PDF Introduction Over the years, many different AHM and uncertainty...