Gas Lift Optimization of an Oil Well using Artificial Neural Networks (ANN)
Keywords:
Machine Learning, Wellhead pressure, Levenberg-Marquardt algorithm, Confusion matrix, Oil well operationsAbstract
This research investigates the gas lift optimization of an oil well using Artificial Neural Networks (ANN). Through statistical analyses of two wells spanning over a two-year period, the critical correlations among wellhead pressure, production rates, and gas compression parameters were reported. Notably, a positive correlation between wellhead pressure and production rate was identified, emphasizing the pivotal role of monitoring and optimizing these operational variables for enhanced efficiency. The research integrates machine learning techniques for gas lift with model recognition and parameter estimation. Leveraging the power of algorithms, like the Levenberg-Marquardt algorithm for the model, the best performance of 0.000000584 was found after 1000 epochs (iterations). The study demonstrates the potential for real-time decision support in oil well operations, offering a pathway for improved responsiveness and adaptability to changing conditions. Practical recommendations derived from the study provide actionable insights for industry practices, facilitating advancements in oil and gas engineering. The validation of machine learning models underscores their reliability in enhancing efficiency and productivity in real-world oil and gas applications. As a culmination of these findings, the research not only advances our understanding of gas lift systems but also provides a roadmap for the implementation of cutting-edge technologies and methodologies in the oil and gas sector.
References
Ahmed A. E, Mohamed, G., Said K., Mohamed E. E. (2021). Prediction of gas-lift performance using neural network analysis. AIMS Energy, 9(2): 355–378.
Ayatollahi, S., Bahadori, A. and Mosh Feghian, A., 2001. “Method Optimizes Aghajari Oil Field Gaslift”, Oil and Gas Journal, Vol. 99, no. 21, pp 47-49.
Beckner, B. & Davidson, J. (2003). Integrated optimization for rate allocation in reservoir simulation. SPEREE, 6(6), 426-432.
Campos, S., Codas, A., Camponogara, E., Gunnerud, V., & Sunjerga, S. (2012). Integrated production optimization of oil fields with pressure and routing constraints: The Urucu field. Computers & Chemical Engineering, 46, 178–189. https://doi.org/10.1016/j.compchemeng.2012.06.016
Clegg, S.M Bucaram and N.W Hein. “Recommendations and Comparisons for Selecting Artificial Lifts Methods”. Paper SPE 24834, Journal of Petroleum Technology, December 1993.
Carlos, B., Jackson, J., Lenin, P., Nicolas, G., Neil,.G., Luis, C. & Edison, P. (2013). Alternative deployment of an ESP system in South America. Paper presented at the SPE artificial lift conference and exhibition, Cartagena, Colombia.
Djikpesse, H., Couet, B., & Wilkinson, D. (2010). Gas lift optimization under facilities constraints. Paper presented at the 24th Annual SPE International Conference and Exhibition, Tinipa-Calabar, Nigeria.
Hisham, K.B. M. and Vincent, G. (2017). Evaluation of continuous gas lift systems in an oil well. International Journal of Petroleum and Petrochemical Engineering. Vol 3, Issue 3, PP 66-75.
Imran, A. H, Sarfraz, A. J., Khalil R. M., Sohail, N. and Khair. J. B. (2018). Design and Economic Evaluation of the ESP and Gas Lift on the Dead Oil Well. International Journal of Current Engineering and Technology, Vol.8, No.6, 1548 - 1553.
Kosmidis, V., Perkins, J. & Pistikopoulos, E. (2004) Optimization of well oil rate allocation in petroleum fields. Industrial & Engineering Chemiistry Research, 43(14), 3513-3527.
Ray, T. & Sarker, R. (2006). Multiobjective Evolutionary Approach to the Solution of Gas Lift Optimization Problems. IEEE Congress on Evolutionary Computation, 3182-3188.
Wang, P. (2003). Development and Application of Production Optimization Techniques for Petroleum Fields. PhD dissertation, Stanford, California: Stanford University.
Okorocha, I. T., Chinwuko, C. E., Mgbemena, C. O., Godfrey, O. C., & Mgbemena, C. E. (2022). Production optimization using gas lift incorporated with artificial neural network. UNIZIK Journal of Engineering and Applied Sciences, 21(1), 842-858.
Okorocha, I. T., Chinwuko, C. E., Mgbemena, C. E., & Mgbemena, C. O. (2020). Gas lift optimization in the oil and gas production process: a review of production challenges and optimization strategies. International Journal of Industrial Optimization, 1(2), 61.
