Neural Network Prediction of Inflow Performance Relationship of Producing Wells- Yemeni Oil Field | ||||
| Journal of Petroleum and Mining Engineering | ||||
| Volume 27, Issue 1 - Serial Number 104, July 2025, Page 7-16 PDF (1.87 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/jpme.2025.339180.1218 | ||||
 View on SCiNiTO | ||||
| Author | ||||
Ghareb hamada 
| ||||
| 277 king faisal st. | ||||
| Abstract | ||||
| The inflow performance relationship (IPR) characterizes the way of behaving of the flowing pressure of the well and the production flow rate, that is a key tool for understanding the reservoir and the well way of behaving and quantifying the production flow rate. The inflow performance relationship is oftentimes needed for design the completion of the well, production optimization of the well, analysis of the nodal determinations, and artificial lift design. Today, various inflow performance relationship correlations subsist in the industry of petroleum, as well as some analytical equations, that generally pain from bound applicability due to high absolute error. In this paper the most correlations were evaluated for prediction IPR for Yemeni oil fields, evaluation between different other correlations. This study presents an analytical method for improved oil flow rate for Yemeni oil wells employing machine learning using input production parameters. New Artificial Neural Networks model was examined by real Yemeni data and gives the best results are obtained from new model and based on the results obtained with AAPE of 0.98 %, R2 of 0.9997 and Standard Deviation of 0.85 compared with AAPE of 8.6 % , R2 of 0.996 and Standard Deviation of 3.55 for Khadafy. A M et al. model which is the best correlation exist, it is recommended to use the developed model to predict IPR. The developed model will be of significant assistance to petroleum industry operators in the Yemeni oil for quick effective estimates of oil flow rates | ||||
| Keywords | ||||
| New model; Yemen field; Artificial Neural Networks; Production test; production correlations | ||||
|
Statistics Article View: 165 PDF Download: 60 |
||||
