May 7, 2024

Application of Machine Learning and Deep Learning to Predict Production Rate of Sucker Rod Pump Wells

View Full Paper

Ask AI

Bookmark

Authors

STS. ThabetMethodist Hospital HZHamdy ZidanAl-Azhar University AEAsmaa ElhadidyNile University

Discussion

Member takes

Implication

Machine learning analysis demonstrates accurate surface fluid rate prediction in sucker rod pump wells, suggesting a viable alternative to physical multiphase flowmeters.

Key Points

Surface fluid production rates can be reliably predicted using gradient boosting and related algorithms across sucker rod pump wells under diverse operating configurations.
A mean absolute percent error of 3.4% and an R2 of 0.937 were achieved by top algorithms, including random forest, when evaluated against real oil well separator measurements.
Assessment of nine machine learning models, including support vector machines, utilized 8,372 data records across three years alongside repeated k-fold cross-validation.
May enable continuous performance monitoring and rapid operational troubleshooting across sucker rod pump wells, serving as an affordable alternative to multiphase flowmeters.

Cite This Study

Thabet et al. (2024) studied this question.

synapsesocial.com/papers/68e6b2a9b6db6435876344da https://doi.org/10.2118/219231-ms

View Full Paper

Ask AI

Bookmark

Implication

Machine learning analysis demonstrates accurate surface fluid rate prediction in sucker rod pump wells, suggesting a viable alternative to physical multiphase flowmeters.

Application of Machine Learning and Deep Learning to Predict Production Rate of Sucker Rod Pump Wells

Discussion

Member takes

Implication

Key Points

Cite This Study

Application of Machine Learning and Deep Learning to Predict Production Rate of Sucker Rod Pump Wells

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider