| Abstrakt: |
The petroleum industry is becoming increasingly concerned with sand production, which is common in the Niger Delta and presents technical, operational, and financial difficulties. Literature has devoted a lot of time to the creation of sanding prediction tools and efficient management techniques. However, the majority of the published theoretical models have been supported by data from other petroleum regions than the Niger Delta or evidence from laboratories. An easy-to-use machine learning model was created that takes into account the idea of dimensionless quantities related to sanding. The parameters taken into consideration included Reynold's Number, Loading factor, Gas-Liquid Ratio and water-cut. An equation for predicting sand production rate is the output from this work. Model validation was carried out on the developed model and the deviation was less than 5% demonstrating the validity of the proposed model. The developed model, when compared to existing models, forecasts superior outcomes, particularly when the boost factor and GLR are very high. The study has a wide range of practical implications, including reservoir management, general well completion design, and field development plans and economics. [ABSTRACT FROM AUTHOR] |
