Research on oil and gas production prediction process based on machine learning

Liu, Zhenzhi; Li, Sanshan; Li, Lü

doi:10.54097/ije.v2i2.7773

Cited by 1 publication

(2 citation statements)

References 21 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The stress sensitivity coefficients were obtained from the core experiments in the study area, and the experimental results showed that the correlation between stress sensitivity coefficients and the permeability was good, which can be expressed by Equation (17).…”

Section: Modeling Of Production Capacity Of Vertical Wells Containing...mentioning

confidence: 99%

“…The third is the analytical or semi-analytical capacity model [13][14][15] based on physical assumptions, which is derived with rigorous ideas, but the current model cannot reasonably characterize the actual seepage characteristics of heavy oil reservoirs, and it is more difficult to apply at the mine site. The fourth is the simulation-based method [16,17], usually using numerical simulation software to predict the production capacity. This method can consider the effects of multiple factors on the production capacity, but it depends on the computational power, model accuracy, and input data accuracy, which is relatively cumbersome.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

New Method for Capacity Evaluation of Offshore Low-Permeability Reservoirs with Natural Fractures

Wang,

Xie,

Liu

et al. 2024

Processes

View full text Add to dashboard Cite

In recent years, the development of two offshore low-permeability oil fields has revealed unexpected challenges. The actual productivity of these fields significantly deviates from the designed capacity. Some wells even outperform the expectations for low-permeability limestone fields. This discrepancy primarily stems from a lack of accurate understanding of natural fractures before and after drilling, resulting in substantial errors in capacity assessment. This paper addresses these challenges by proposing a new production capacity model and evaluation method for both vertical and horizontal wells in low-permeability limestone reservoirs. The method leverages logging curve data, incorporating vertical gradation and fractal analysis to effectively represent the fracture’s complexity and connectivity. It uniquely considers factors such as fracture fractal dimensions, threshold pressure, and stress sensitivity, significantly enhancing prediction accuracy. Furthermore, by analyzing the longitudinal gradient in logging curves, the method effectively identifies strong heterogeneity, leading to more accurate capacity evaluations in actual fields. The results demonstrate that our model reduces the average prediction error to less than 15%, markedly outperforming traditional methods. Calculation results of the newly developed capacity formula align closely with actual production data and tracer test results, showcasing its practical applicability and potential for widespread use. This study notably advances the evaluation of reasonable production capacity in similar offshore reservoirs.

show abstract

Section: Modeling Of Production Capacity Of Vertical Wells Containing...mentioning

confidence: 99%