Production Forecasting: Optimistic and Overconfident—Over and Over Again

Bratvold, Reidar Brumer; Mohus, Erlend; Petutschnig, David; Bickel, E.

doi:10.2118/195914-pa

Cited by 10 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The oilfield development process is intricate, and real production data often display a wide range of values across dimensions, with uneven temporal distribution causing increased noise levels . These traits not only present challenges in model training but also induce model instability, impacting its robustness.…”

Section: Methodsmentioning

confidence: 99%

Well Production Forecast Post-Liquid Lifting Measures: a Transformer-Based Seq2Seq Method with Attention Mechanism

Lu,

Cui,

Qian

et al. 2024

Energy Fuels

View full text Add to dashboard Cite

Oilfield production forecasting plays a crucial role in petroleum production. However, traditional methods, such as the Arps decline curve, merely reflect the trend of production history and fail to comprehensively consider the impact of measurement adjustments on production. In contrast, numerical simulation methods, although capable of comprehensively considering various factors, face practical application limitations due to their complexity and high computational costs. This study proposes a method for oil well production forecasting that integrates the attention mechanism with the Seq2Seq structure. Leveraging data from multiple wells in the ultrahigh water cut phase of the SL oilfield in China, a learning sample data set is established using dynamic time warping (DTW) for feature sorting. A production forecasting model, a transformer-based Seq2Seq deep learning algorithm capable of considering the impact of liquid lifting measures, is established. Comprehensive testing validates the superior performance of the model, with coefficients of determination on the training, validation, and testing sets being 0.9730, 0.9649, and 0.9461, respectively, clearly outperforming the traditional long−short-term memory (LSTM) model. Taking nine sample wells as an example, the model predicts oil production for the next 12 months under different liquid lifting magnitudes and determines the optimal liquid lifting magnitudes and corresponding monthly average incremental oil levels with the aid of a genetic algorithm. Experimental results align well with the empirical trends of the target oilfield, demonstrating that this model can be used as an effective tool for predicting well production and conducting analysis of the potential of liquid lifting. By optimizing the liquid lifting magnitudes of oil wells through accurate oil production predictions, the model enhances intelligent decision-making and optimizes production strategies, providing significant practical value in actual oilfield management.

show abstract

Section: Methodsmentioning

confidence: 99%

Well Production Forecast Post-Liquid Lifting Measures: a Transformer-Based Seq2Seq Method with Attention Mechanism

Lu,

Cui,

Qian

et al. 2024

Energy Fuels

View full text Add to dashboard Cite

show abstract

“…To increase the expected reward, data acquisition needs to be observable, relevant, material, and economic [76]. The distance-based generalized sensitivity analysis, as proposed by Fenwick et al [77], proves instrumental in showing the sensitivity of parameters, including their interdependencies, on NPV.…”

Section: Sensitivity Analysis and Data Acquisition Actionsmentioning

confidence: 99%

Artificial Intelligence-Centric Low-Enthalpy Geothermal Field Development Planning

Clemens,

Chiotoroiu,

Corso

et al. 2024

Energies

View full text Add to dashboard Cite

Low-enthalpy geothermal energy can make a major contribution towards reducing CO2 emissions. However, the development of geothermal reservoirs is costly and time intensive. In particular, high capital expenditures, data acquisition costs, and long periods of time from identifying a geothermal resource to geothermal heat extraction make geothermal field developments challenging. Conventional geothermal field development planning follows a linear approach starting with numerical model calibrations of the existing subsurface data, simulations of forecasts for geothermal heat production, and cost estimations. Next, data acquisition actions are evaluated and performed, and then the models are changed by integrating the new data before being finally used for forecasting and economics. There are several challenges when using this approach and the duration of model rebuilding with the availability of new data is time consuming. Furthermore, the approach does not address sequential decision making under uncertainty as it focuses on individual data acquisition actions. An artificial intelligence (AI)-centric approach to field development planning substantially improves cycle times and the expected rewards from geothermal projects. The reason for this is that various methods such as machine learning in data conditioning and distance-based generalized sensitivity analysis assess the uncertainty and quantify its potential impact on the final value. The use of AI for sequential decision making under uncertainty results in an optimized data acquisition strategy, a recommendation of a specific development scenario, or advice against further investment. This approach is illustrated by applying AI-centric geothermal field development planning to an Austrian low-enthalpy geothermal case. The results show an increase in the expected value of over 27% and a reduction in data acquisition costs by more than 35% when compared with conventional field development planning strategies. Furthermore, the results are used in systematic trade-off assessments of various key performance indicators.

show abstract

“…In the original paper, Keelin (2016) used three term metalogs to model expert elicited probability distributions over 259 financial assets as part of a bidding decision analysis and showed that use of metalog's resulted in a better bidding decision compared to modeling theses same uncertainties with three-branch discretization, a widely accepted and commonly used modelling assumption in decision analysis. Bratvold, et. al.…”

Section: Eliciting Expert Opinionmentioning

confidence: 99%

The Metalog Distributions: Virtually Unlimited Shape Flexibility, Combining Expert Opinion in Closed Form, and Bayesian Updating in Closed Form

Keelin¹,

Howard²

2021

Preprint

View full text Add to dashboard Cite

Users of probability distributions frequently need to convert data (empirical, simulated, or elicited) into a continuous probability distribution and to update that distribution when new data becomes available. Often, it is unclear which traditional probability distribution(s) to use, fitting to data is laborious and unsatisfactory, little insight emerges, and updating with Bayes rule is impractical. Here we offer an alternative -- a family of continuous probability distributions, fitting methods, and tools that: provide sufficient shape and boundedness flexibility to closely match virtually any probability distribution and most data sets; involve a single set of simple closed-form equations; stimulate potentially valuable insights when applied to empirical data; are simply fit to data with ordinary least squares; are easy to combine (as when weighting the opinion of multiple experts), and, under certain conditions, are easily updated in closed form according to Bayes rule when new data becomes available. The Bayesian updating method is presented in a way that is readily understandable as a fisherman updates his catch probabilities when changing the river on which he fishes. While metalog applications have been shown to improve decision-making, the methods and results herein are broadly applicable to virtually any use of continuous probability in any field of human endeavor. Diverse data sets may be explored and modeled in these new ways with freely available spreadsheets and tools.

show abstract

Production Forecasting: Optimistic and Overconfident—Over and Over Again

Cited by 10 publications

References 13 publications

Well Production Forecast Post-Liquid Lifting Measures: a Transformer-Based Seq2Seq Method with Attention Mechanism

Well Production Forecast Post-Liquid Lifting Measures: a Transformer-Based Seq2Seq Method with Attention Mechanism

Artificial Intelligence-Centric Low-Enthalpy Geothermal Field Development Planning

The Metalog Distributions: Virtually Unlimited Shape Flexibility, Combining Expert Opinion in Closed Form, and Bayesian Updating in Closed Form

Contact Info

Product

Resources

About