Deriving Alkali Polymer Parameter Distributions from Core Flooding by Applying Machine Learning in a Bayesian Framework to Simulate Incremental Oil Recovery
- Dominik Steineder (OMV E&P) | Gisela Vanegas (OMV E&P) | Torsten Clemens (OMV E&P) | Markus Zechner (OMV E&P/Stanford University)
- Document ID
- Society of Petroleum Engineers
- SPE Europec featured at 82nd EAGE Conference and Exhibition, 8-11 December, Amsterdam, The Netherlands
- Publication Date
- Document Type
- Conference Paper
- 2020. Society of Petroleum Engineers
- 7.6.6 Artificial Intelligence, 5.5.2 Core Analysis, 1.6 Drilling Operations, 1.6.9 Coring, Fishing
- Bayesian Framework, Machine Learning, Principal Component Analysis, Multiple Objective Functions, Alkali Polymer Flooding
- 4 in the last 30 days
- 27 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 28.00|
Various physico-chemical processes are affecting Alkali Polymer (AP) Flooding. Core floods can be performed to determine ranges for the parameters used in numerical models describing these processes. Because the parameters are uncertain, prior parameter ranges are introduced and the data is conditioned to observed data. It is challenging to determine posterior distributions of the various parameters as they need to be consistent with the different sets of data that are observed (e.g. pressures, oil and water production, chemical concentration at the outlet).
Here, we are applying Machine Learning in a Bayesian Framework to condition parameter ranges to a multitude of observed data.
To generate the response of the parameters, we used a numerical model and applied Latin Hypercube Sampling (2000 simulation runs) from the prior parameter ranges.
To ensure that sufficient parameter combinations of the model comply with various observed data, Machine Learning can be applied. After defining multiple Objective Functions (OF) covering the different observed data (here six different Objective Functions), we used the Random Forest algorithm to generate statistical models for each of the Objective Functions.
Next, parameter combinations which lead to results that are outside of the acceptance limit of the first Objective Function are rejected. Then, resampling is performed and the next Objective Function is applied until the last Objective Function is reached. To account for parameter interactions, the resulting parameter distributions are tested for the limits of all the Objective Functions.
The results show that posterior parameter distributions can be efficiently conditioned to the various sets of observed data. Insensitive parameter ranges are not modified as they are not influenced by the information from the observed data. This is crucial as insensitive parameters in history could become sensitive in the forecast if the production mechanism is changed.
The workflow introduced here can be applied for conditioning parameter ranges of field (re-)development projects to various observed data as well.
|File Size||1008 KB||Number of Pages||21|
Almeida, F.la R.; Formentin, H.N.; Machio, C.; Davolio, A. and D.J. Schiozer. 2018. Influence of Additional Objective Functions on Uncertainty Reduction and History Matching. Paper SPe-190804 presented at the SPE Europec featured at 80th EAGE Conference and Exhibition. Copenhagen. Denmakr. 11-14 June 2018.
Al-Shuaili, K.; Svec, Y.; Guntupalli, S.; Al-Amri, M.; Al-Hinai, G.; Mohammed, S.; Al-Hadhrami, H.; van Batenburg, D.W. and S. de Kruijif. 2018. Field Piloting of Alkaline Surfactant Polymer in Sultanate of Oman. Paper SPE 190392 presented at the SPE EOR Conference at Oil and Gas West Asia. Muscat. Oman. 26-28 March 2018.
Google, 2019.http://developers.google.com/machine-leaming/glossary. Accessed 9 January 2020.
Hutahaean, J.J.; Demyanow, V. and M.A. Christie. 2015. Impact of Model Parameterisation and Objective Choices on Assisted History Matching and Reservoir Forecasting. Paper SPE 176389 presented at the SPE/IATMI Asia Pacific Oil and Gas Conference and Exhibition. Nusa Dua. Indonesia. 20-22 October 2015.
Klie, H., Klie, A., Rodriguez, A., Monteagudo, J., Primera, A. and M. Quesada. 2019. Improving Field Development Decisions in the Vaca Muerta Shale Formation by Efficient Integration of Data, Al and Physics. Paper URTEC: 938 presented at the Unconventional Resources Technology Conference. Denver. USA. 11-24 July 2019.
Nghiem, L.; Skoreyko, F.; Gorucu, S.E.; Dang, C. and V. Shrivastava. 2012. A Framework for Mechanistic Modeling of Alkali-Surfactant-Polymer Process in an Equation-of-State Compositional Simulator. Paper SPE 182628 presented at the SPE Reservoir Simulation Conference. Montgomery. USA. 20-22 February 2017.
Schulze-Riegert, R.; Daniali, A.; Nwakile, M.; Selberg, S.; Skripkin, W.; Chugunov, N. and J. Carter. 2017. Data Analysis Used in Multiple-Realization Workflows for History Matching - A North Sea Case Study. Paper SPE 185877 presented at the SPE EUROPEC featured at the 79th EAGE Conference and Exhibition. Paris. France. 12-15 June 2017.
Volokitin, Y.; Shuster, M.; Karpan, V.; Koltsov, I.; Mikhaylenko, E.; Bondar, M.; Podberezhny, M.; Rakitin, A.; van Batenburg, D.W.; Parker, A.R.; de Kruijf, S.; Southwick, J.G.; de Reus, J.; van den Pol, E.; vand der Heyden, F.H.J.; Boels, L.; Weyer, D.A.Z. and M. Brewer. 2018. Results of Alkaline-Surfactant-Polymer Flooding Pilot at West Salym Field. Paper SPE 190382 presented at the SPE EOR Conference at Oil and Gas West Asia. Muscat. Oman.26-28 March 2018.
Xiao, R.; Teletzke, G.F.; Lin, M.W.; Glotzbach, R.C.; Aitkulov, A.; Yeganeh, M.; Jaishankar, A. and J. Hegner. 2017. A Novel Mechanism of Alkaline Flooding to Improve Sweep Efficiency for Viscous Oils. Paper SPE 187366 presented at the SPE Annual Technical Conference and Exhibition. San Antonio. USA. 9-11 October 2017.