Advances in Parallel Reservoir Simulation Technology, Using Multi-Threaded Algorithms
- Sina Mohajeri (Engineering Support & Technology Development Company) | Reza Eslahi (Engineering Support & Technology Development Company) | Maryam Bakhtiari (Engineering Support & Technology Development Company) | Mostafa Zeinali (Engineering Support & Technology Development Company) | Hamed Rajabi (Engineering Support & Technology Development Company) | Ali Alizadeh (Engineering Support & Technology Development Company) | Ebrahim Sharifi (Engineering Support & Technology Development Company) | Emad Mortezazadeh (Engineering Support & Technology Development Company)
- Document ID
- Society of Petroleum Engineers
- Abu Dhabi International Petroleum Exhibition & Conference, 12-15 November, Abu Dhabi, UAE
- Publication Date
- Document Type
- Conference Paper
- 2018. Society of Petroleum Engineers
- 1.6.9 Coring, Fishing, 5.5 Reservoir Simulation, 2 Well completion, 2.7.1 Completion Fluids, 1.6 Drilling Operations, 5 Reservoir Desciption & Dynamics, 2.7 Completion Fluids
- Shared Memory Systems, Parallel Reservoir Simulation, Multi-Threaded Algorithms, Linear Systems, Fractured Reservoirs
- 3 in the last 30 days
- 35 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 8.50|
|SPE Non-Member Price:||USD 25.00|
The popularity of parallel reservoir simulation has increased in recent years due to availability of multi-core CPUs and the opportunity for run-time reduction, especially in complex and fractured reservoir models that include large amounts of data. However, the accuracy of classical domain decomposition techniques significantly decreases as the number of processing units increases. The scope of this work is mainly concentrated on development of multi-threaded algorithms to overcome the mentioned problem in parallel reservoir simulations. In this work, at first, a newly developed, parallel, three dimensional, fully implicit, three-phase black-oil reservoir simulator is introduced. For solving large scale linear systems produced in giant black-oil models, which are probable cases in real world reservoir models, BiCG-Stabilized linear solver, enhanced with various preconditioners including, Blocked Incomplete LU factorization, Constrained Pressure Residual using Algebraic Multi-Grid as its pressure smoother is constructed with a variety of parallelization techniques applied in each part. The presented multi-threaded algorithms can utilize any user-defined number of CPU cores. The algorithm has been evaluated by SPE10 with 1122000 grid blocks and a real data file containing 277875 grid blocks. The result indicates that increasing the number of CPU threads does not have any negative effect on the simulation performance in case of high memory bandwidth limits. For SPE10 data file, 1.70, 2.4, and 2.3 speedup ratios (in comparison with a single thread) has been observed for 2, 4, and 6 parallel threads, respectively. In the real case, the simulation speed is increased by a factor of 1.49, 1.74, and 1.75 for 2, 4, and 6 parallel threads, respectively. Using multi-threaded algorithms as a novel approach in parallel simulations, the solution configuration does not change from the single core case and follows the same exact path. This however is not the case for other simulators that use domain-decomposition techniques. These techniques insert numerical errors in the solution of linear systems, which lead to convergence problems in the whole system.
|File Size||888 KB||Number of Pages||9|
Liu, H., Wang, K., Chen, Z., Luo, J., Wu, S., & Wang, B. (2015, September 14). Development of Parallel Reservoir Simulators on Distributed-memory Supercomputers. Society of Petroleum Engineers. doi:10.2118/175573-MS
Meijerink, J. A.; Vorst, Van Der; A, H. (1977). "An iterative solution method for linear systems of which the coefficient matrix is a symmetric M-matrix". Mathematics of Computation. 31 (137): 148–162. doi:10.1090/S0025-5718-1977-0438681-4. ISSN 0025-5718.
Chapman, B. and LaGrone, J., "OpenMP," in: Encyclopedia of Parallel Computing, ed. By D. Padua, Boston, MA: Springer US, 2011, pp. 1365–1371, url: http://dx.doi. org/10.1007/978-0-387-09766-4
Blaise, B., "Introduction to Parallel Computing," in: Lawrence Livermore National UCRLMI-133316 (EC3500 2015), url: https://computing.llnl.gov/tutorials/parallel_comp/.