|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
The complete paper explores the use of multilevel derivative-free optimization for history matching, with model properties described using principal component analysis (PCA) -based parameterization techniques. This paper describes an accurate, three-step, machine-learning-based early warning system that has been used to monitor production and guide strategy in the Shengli field. This paper discusses how machine learning by use of multiple linear regression and a neural network was used to optimize completions and well designs in the Duvernay shale.
The industry increasingly relies on forecasts from reservoir models for reservoir management and decision making. However, because forecasts from reservoir models carry large uncertainties, calibrating them as soon as data come in is crucial. The complete paper explores the use of multilevel derivative-free optimization for history matching, with model properties described using principal component analysis (PCA) -based parameterization techniques. The results of the authors’ research showed promising benefits from the use of a systematic procedure of model diagnostics, model improvement, and model-error quantification during data assimilations. A challenging problem of automated history-matching work flows is ensuring that, after applying updates to previous models, the resulting history-matched models remain consistent geologically.
Monte Carlo simulation is a process of running a model numerous times with a random selection from the input distributions for each variable. The results of these numerous scenarios can give you a "most likely" case, along with a statistical distribution to understand the risk or uncertainty involved. Computer programs make it easy to run thousands of random samplings quickly. Monte Carlo simulation begins with a model, often built in a spreadsheet, having input distributions and output functions of the inputs. The following description is drawn largely from Murtha.
If least-squares linear regression is used to compute N in Step 5, an equation analogous to Eq. 17 is used (where Eow is substituted for Eowf). This solution method is iterative because the material-balance error must be minimized. This calculation is carried out with a trial-and-error method or a minimization algorithm. Least-squares linear regression and minimization algorithms have become standard features in commercial spreadsheets.
An understanding of statistical concepts is important to many aspects of petroleum engineering, but especially reservoir modeling and simulation. The discussion below focuses on a range of statistical concepts that engineers may find valuable to understand. The focus here is classical statistics, but differences in the application for geostatistics are included. A quantitative approach requires more than a headlong rush into the data, armed with a computer. Because conclusions from a quantitative study are based at least in part on inferences drawn from measurements, the geoscientist and reservoir engineer must be aware of the nature of the measurement systems with which the data are collected. Each of these scales is more rigorously defined than the one before it. The nominal and ordinal scales classify observations into exclusive categories.
Measured production data always has a degree of uncertainty which is not considered generally. Uncertainty in rates is defined as the size of its margin of doubt. It's crucial to characterize this uncertainty for various reasons; Reporting accurate rates while monitoring projects leads to better management decisions. It will give us a practical tool to evaluate the quality of a given measurement technology applied. Finally, if any operating condition is changed in a mature high water cut field like Estancia Cholita, it's crucial to determine oil percentage to evaluate the new operations success. Oil rates are usually quantified by measuring total liquid rates with test separator and then taking a sample to determine the oil percentage. There are so many factors causing uncertainty in these procedures (duration of the oil sampling, flow regimes, skills of the operator performing the measurement, etc.) that is difficult to fully describe error from a theoretical point of view.
A new comprehensive uncertainty method is proposed. First, a table of error as a function of water cut is presented by using error propagation theory. Second, real field data from forty different wells was analyzed. A well-defined period without any changes in operating conditions is chosen per well and an Arps decline curve is fitted. Then a histogram was created to establish an error distribution function (the error is defined as the difference between the measured rate and fitted curve prediction which is taken as the real state for calculation purposes). Finally, a correspondence between error and confidence interval was highlighted to repeated or monitored specifically. A new frequency proposal and test priority is presented. The wells are classified into type I type II and type II according to the relationship between flow and uncertainty. This classification is key to define a new measurement prioritization system
Impact of workovers and interventions in high water cut wells, like Estancia Cholita, are now monitored accurately. This allows improved control over production leading to better decision making. Specially, when monitoring a pilot, such as polymer injection where the accuracy of the devices used becomes very useful. Frequency and prioritization of rate measurements are specifically described for individual wells.
Using this novel approach, unnecessary measurements and operations are reduced. Having clean data could lead to a successful data mining analysis. Workovers and intervention impact on high water cut wells can be monitored more accurately. By knowing the error in the production rates, future projects will be well defined managed and evaluated.
Machado de Almeida Duque, Maria Clara (Universidade Federal do Rio de Janeiro) | Souza Chaves, Gabriela (Universidade Federal do Rio de Janeiro) | de Oliveira Monteiro, Danielle (Universidade Federal do Rio de Janeiro) | Velasco Medani, Luciana (Universidade Federal do Rio de Janeiro) | Martins Ferreira Filho, Virgílio José (Universidade Federal do Rio de Janeiro)
Well production in oil fields is a dynamic and complex activity. The patterns and characteristics inherent to the well, such as pressures and flow rates, are changing based on production time and the fluid composition – a complex multiphase mixture composed of oil, water, and gas. Thus, it is necessary to evaluate well behavior with periodic production tests. This paper proposes an automatic tool based on machine learning models to assist the production tests validation process in a quick manner. The developed methodology was applied to 13 representative wells of a Brazilian offshore oil field. For each examined well, a dataset is created with operation variables obtained from valid and invalid production tests. Six classification algorithms are analyzed, Logistic Regression, Naïve Bayes classifier, K-Nearest Neighbor, Decision Tree, Random Forest and Support Vector Machine (SVM) in reason to automatically label a new production test as valid or invalid, according to production historical data for the well. The dataset was divided into training and validation sets. The training set was used to perform feature selection, to calibrate and choose the proper model. The validation set was then used at the end of the procedure to evaluate obtained results, by comparing the model’s output with real test labels. From the results obtained in the case study, it was possible to identify that IGLR (Injection Gas/Liquid Ratio), oil flow rate and the pressure loss between wellhead and platform were representatives for most of the wells, which implies that these variables have a huge influence at the production well test validation. Furthermore, the validation set indicates that SVM and logistic regression were the models with the best performance. Besides that, accurate results were achieved, since the model correctly classified at least 5 of the 6 tests in 70% of wells analyzed, and for the remaining wells, 4 of 6 production tests.
Temizel, Cenk (Saudi Aramco) | Canbaz, Celal Hakan (Ege University) | Gok, Ihsan Murat (NESR) | Roshankhah, Shahrzad (California Institute of Technology) | Palabiyik, Yildiray (Istanbul Technical University) | Deniz-Paker, Melek (Independent Consultant) | Hosgor, Fatma Bahar (Petroleum Software LLC) | Ozyurtkan, Mustafa Hakan (Istanbul Technical University) | Aksahan, Firat (Ege University) | Gormez, Ender (Middle East Technical University)
As major oil and gas companies have been investing in shale oil and gas resources, even though has been part of the oil and gas industry for long time, shale oil and gas has gained its popularity back with increasing oil prices. Oil and gas industry has adapted to the low-cost operations and has started investing in and utilizing the shale oil sources significantly. In this perspective, this study investigates and outlines the latest advances, technologies, potential of shale oil and gas reservoirs as a significant source of energy in the current supply and demand dynamics of oil and gas resources. A comprehensive literature review focusing on the recent developments and findings in the shale oil and gas resources along with the availability and locations are outlined and discussed under the current dynamics of the oil and gas market and resources. Literature review includes a broad spectrum that spans from technical petroleum literature with very comprehensive research using SCOPUS database to other renowned resources including journals and other publications. All gathered information and data are summarized.Not only the facts and information are outlined for the individual type of energy resource but also the relationship between shale oil/gas and other unconventional resources are discussed from a perspective of their roles either as a competing or a complementary source in the industry. In this sense, this study goes beyond only providing raw data or facts about the energy resources but also a thorough publication that provides the oil and gas industry professional with a clear image of the past, present and the expected near future of the shale oil/gas as it stands with respect to other energy resources. Among the few existing studies that shed light on the current status of the oil and gas industry facing the rise of the shale oil are up-to-date and the existing studies within SPE domain focus on facts only lacking the interrelationship between heavy and light oil as a complementary and a competitor but harder-to-recover form of hydrocarbon energy within the era of rise of renewables and other unconventionals. This study closes the gap and serves as an up-to-date reference for industry professionals. 2 SPE-198994-MS
Mayorga Cespedes, Edgar Alberto (Ecopetrol) | Roostaei, Morteza (RGL Reservoir Management Inc.) | Uzcátegui, Alberto A. (RGL Reservoir Management Inc.) | Soroush, Mohammad (RGL Reservoir Management Inc., University of Alberta) | Izadi, Hossein (University of Alberta) | Hosseini, Seyed Abolhassan (RGL Reservoir Management Inc., University of Alberta) | Schroeder, Brad (RGL Reservoir Management Inc.) | Mahmoudi, Mahdi (RGL Reservoir Management Inc.) | Gomez, Dionis M. (Ecopetrol) | Mora, Edgar (Ecopetrol) | Alpire, Javier (Ecopetrol) | Torres, Joselvis (Ecopetrol) | Fattahpour, Vahidoddin (RGL Reservoir Management Inc.)
Designing/Selecting the proper sand control mechanism for horizontal wells in unconsolidated heavy-oil reservoirs tend to be under-looked in some cases. Stand-alone completions pose some sand control challenges, which could jeopardize the oil production or even lead to critical problems. Massive sand production, screen/formation plugging, hot-spots and mechanical integrity failures are some of the well-known issues. This study attempts to optimize the sand control design for horizontal wells in a heavy-oil field in Colombia.
A careful selection of representative core data was made to study the variation of sand particle size distribution (PSD) within the development area. Reservoir fluid properties were analyzed. Based on PSD variation and current design criteria in the industry, several seamed slotted-liner configurations were proposed as an alternative completion for testing. Later, a series of large-scale sand retention tests (SRTs and FSTs) were performed to assess the selected alternatives under typical field production conditions. Effects of aperture size and open flow area (OFA) were investigated to evaluate flow and sand control performance.
This investigation started by a detailed study of the PSD, particle shape variation and composition of fines in the development area. The PSD then classified into four distinct minor and major sand facies, ranging from medium to very coarse sand with different fines content. Further Investigations have shown that current design is only suitable for a limited number of the PSDs, while the overall PSD classes indicate the requirement for wider slot aperture sizes. The results of the SRTs indicated that the flow performance of the screen is mainly controlled by the slot aperture. Choosing the optimized aperture size avoids unacceptable sanding even for the multiphase flow scenarios with gas. Results also indicated that by increasing the aperture size and application of the seamed slots for the studied formation, plugging could be mitigated. Finally, a detailed finite element analysis was conducted to compare the current slotted liner design and the optimized design based on the experimental testing.
A comprehensive sand control design workflow for cold primary heavy oil production in horizontal wells is presented in this work. The current study is one of the first that investigates and compares conventional straight slotted liners with seamed slotted liners at larger scale for this field. Moreover, this study helps to better understand the effect of design parameters of seamed slotted liners on sand control, flow performance and mechanical strength.