Related references
Note: Only part of the references are listed.
Article
Energy & Fuels
Anand Korde et al.
Summary: This study develops a probabilistic methodology for decline curve analysis in unconventional reservoirs using Bayesian model-fitting algorithms and deterministic models, with the aim of accurate production forecasting and uncertainty quantification. Different sampling vehicles, including Gibbs sampling, Metropolis-Hastings algorithm, and approximate Bayesian computation, are employed to sample parameter values from posterior distributions and estimate DCA parameter prediction intervals. Testing on over 74 oil and gas wells shows that the DCA-Bayesian models are well calibrated, resulting in low production errors and narrow uncertainty bounds for production history data sets.
SPE RESERVOIR EVALUATION & ENGINEERING
(2021)
Article
Engineering, Petroleum
Boxiao Li et al.
Summary: Decline curve analysis is widely used in production forecasting for unconventional hydrocarbon reservoirs, but traditional methods have limitations. Machine-learning methods are being used as alternatives, with dynamic production rescaling (DPR) developed to improve accuracy. DPR has shown significant error reduction across different basins and formations, proving its effectiveness and efficiency.
Article
Energy & Fuels
Xiangwen Kong et al.
Summary: In this study, a comprehensive numerical compositional model combined with EDFM method was used to evaluate the geological and engineering controls on gas huff-n-puff in Duvernay shale volatile oil reservoirs. The results showed that fracture conductivity, natural fracture density, period of primary depletion, and natural fracture permeability are the most sensitive parameters for incremental oil recovery from gas huff-n-puff. Low fracture conductivity and a short period of primary depletion could greatly reduce the economic efficiency of the process.
Article
Engineering, Petroleum
R. D. Hazlett et al.
Summary: The new discrete fracture model approach presented in this study addresses some of the challenges in unconventional reservoir evaluation with DCA, demonstrating advantages in accurately characterizing reservoir properties. The method allows for reinterpretation of previous results analyzed by traditional DCA and enables classification of new type curves.
Article
Energy & Fuels
Zekun Guo et al.
Summary: A production prediction model based on machine learning technique was developed for horizontal wells targeting the Duvernay Formation, identifying the most important factors influencing production. The model performance was evaluated using statistical models, with the GPR model showing the best performance. Sensitivity analysis indicated that certain factors, such as condensate-gas ratio and fluid volume, were crucial for predicting cumulative oil production.
Article
Engineering, Petroleum
Kittiphong Jongkittinarukorn et al.
