tNavigator Integrated Solutions
tNavigator is the first integrated-by-design reservoir management software – connecting interpretation, geomodeling, PVT, well modeling, simulation and surface networks within one interface.
By reducing the need for 6 software programs down to 1, all data and knowledge stays in a single platform, increasing performance and efficiency. tNavigator’s integrated solution offers unrivaled simulation speed and scalability – for game changing, integrated affordable reservoir management.
Because when all of the pieces come together, performance is transformative.
Integration
- Workflow based integrated reservoir modeling: geological model + dynamic model + wellbore model + PVT + surface networks + history matching + uncertainty analysis
- Seamless integration across the workflow
- Benefits include: working with the latest data, intuitive and easy to use
Efficiency
- Fully coupled, fully implicit reservoir simulation and surface network modeling
- Cost efficient: work from 1 interface, not 6
- Time efficiency with CPU and GPU
- Works where you want to work: laptop, workstation, cluster or cloud
Performance
- Ultimate simulation performance with multiple CPU and GPU computing
- CPU/GPU hybrid simulation engine to optimize usage of your computing environment
- GPU can deliver up to 10 times speed-up on model calculation versus CPU only
Insight
- Advanced static modeling with geology designer: seismic interpretation + petrophysics + well correlations + complex faults + advanced grids accelerated with parallel computing
Fully coupled, fully implicit modeling of reservoir and surface network for improved fidelity and speed
- Always working off the latest information
tNavigator’s integrated solution provides insight throughout your workflow, from interpretation to surface network modeling. With full visibility, errors and wasted time are reduced, and a better outcome is achieved.
Interested in learning more about RFD’s tNavigator Integrated Solution? Contact the closest location to you.
tNavigator Modules
Fully Integrated Solution for Reservoir Engineers& Geoscientists
tNavigator is a software package offered as a single executable, allowing the user to build static and dynamic reservoir models, run dynamic simulations, calculate PVT properties of fluids, build surface network models, calculate lifting tables and perform extended uncertainty analysis as a part of one integrated workflow. All the parts of the workflow share a common proprietary internal data storage system, super-scalable parallel numerical engine (tested up to 10240 CPU and 35840 GPU cores with model sizes exceeding 1 billion active grid blocks), data I/O mechanism and graphical user interface.
tNavigator supports third party data input formats. The format converters are embedded into the executable and provide on-the-fly conversion of input data decks into the internal data storage system.
tNavigator licensing is enabled for local and network environments. Local licenses are provided for standalone workstations and laptops and require a USB dongle and corresponding license file. Network licenses for LAN and WAN networks are provided by a license server installed on Linux or Windows computer systems (physical or virtual). The license server requires access to a USB dongle and its license file. The license server is designed for high-load and could provide usage statistics for FlexNet® and OpenIT® monitoring systems.
Geology Designer
Geology Designer allows the user to build a static model from scratch.
Key Features
- Load and edit interpreted seismic surfaces, well trajectories, logs and well picks, facies properties, rock properties, petrophysical information, point sets and other objects. Formats exported from some legacy third-party tools can be loaded to Geology Designer.
- 2D and 3D seismic: import of SEG-Y format in time or depth, creation and visualisation of in-lines, cross-lines and time-slices. Seismic interpretation: time/depth law import. Display of wells, well logs and markers on the Seismic tab (2D).
- Synthetics: calculation to create a synthetic velocity curve, synthetic acoustic impedance, reflection coefficient and interval velocity curves to compare with the seismic data.
- Seismic attributes computation: coherence, instantaneous frequency, amplitude, phase.
- Support for coordinate reference systems by country or EPSG codes.
- 2D and 3D visualisation, histograms, crossplots, vertical proportion curves.
- Well correlation window allows the user to work with many wells at the same time. Automatic and manual well correlation is capable of handling hundreds of wells at the same time. Ability to work with several ghost curves at the same time.
- Faults: loading faults in standard formats, faults creation via polygons, fault editing, build a 3D grid with faults.
- Structural modelling, local grid editing.
- Facies analysis, variograms, property interpolation: least-squares, inverse distance weighing (IDW), kriging, co-kriging, Gaussian simulation (SGS) and multi-point statistic simulation.
- Fluid-in-place in 2D and 3D.
- Calculator to work with all project objects.
- Python based workflows.
- Geosteering.
- All of the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black oil, compositional or thermal), Assisted History Matching and Uncertainty provides the capability to create static and dynamic models in one graphical interface, run simulations, analyse results and carry out assisted history matching and uncertainty analysis.
Combined with the Assisted History Matching & Uncertainty package, the Geology Designer allows the user to capture static uncertainties with full ranges and distributions, offering the ability to carry out a sensitivity and uncertainty analysis on your geological grid, without the need to build a full hydrodynamic model.
- tNavigator Integrated Solutions
- tNavigator Modules
- Geology Designer
- Model Designer
- PVT Designer
- VFP Designer
- Network Designer
- Black Oil Simulator
- Compositional Simulator
- Thermal Compositional Simulator
- Assisted History Matching and Uncertainty analysis
- Graphical User Interface
- tNavigator Technical Description
- Supercomputing in Reservoir Simulation
- tNavigator on workstations
- tNavigator on clusters
- Hardware
tNavigator Integrated Solutions
tNavigator is the first integrated-by-design reservoir management software – connecting interpretation, geomodeling, PVT, well modeling, simulation and surface networks within one interface.
Because when all of the pieces come together, performance is transformativ
Integration
- Workflow based integrated reservoir modeling: geological model + dynamic model + wellbore model + PVT + surface networks + history matching + uncertainty analysis
- Seamless integration across the workflow
- Benefits include: working with the latest data, intuitive and easy to use
Efficiency
- Fully coupled, fully implicit reservoir simulation and surface network modeling
- Cost efficient: work from 1 interface, not 6
- Time efficiency with CPU and GPU
- Works where you want to work: laptop, workstation, cluster or cloud
Performance
- Ultimate simulation performance with multiple CPU and GPU computing
- CPU/GPU hybrid simulation engine to optimize usage of your computing environment
- GPU can deliver up to 10 times speed-up on model calculation versus CPU only
Insight
- Advanced static modeling with geology designer: seismic interpretation + petrophysics + well correlations + complex faults + advanced grids accelerated with parallel computing
Fully coupled, fully implicit modeling of reservoir and surface network for improved fidelity and speed
- Always working off the latest information
tNavigator’s integrated solution provides insight throughout your workflow, from interpretation to surface network modeling. With full visibility, errors and wasted time are reduced, and a better outcome is achieved.
tNavigator Modules
Fully Integrated Solution for Reservoir Engineers& Geoscientists
tNavigator is a software package offered as a single executable, allowing the user to build static and dynamic reservoir models, run dynamic simulations, calculate PVT properties of fluids, build surface network models, calculate lifting tables and perform extended uncertainty analysis as a part of one integrated workflow. All the parts of the workflow share a common proprietary internal data storage system, super-scalable parallel numerical engine (tested up to 10240 CPU and 35840 GPU cores with model sizes exceeding 1 billion active grid blocks), data I/O mechanism and graphical user interface.
tNavigator supports third party data input formats. The format converters are embedded into the executable and provide on-the-fly conversion of input data decks into the internal data storage system.
tNavigator licensing is enabled for local and network environments. Local licenses are provided for standalone workstations and laptops and require a USB dongle and corresponding license file. Network licenses for LAN and WAN networks are provided by a license server installed on Linux or Windows computer systems (physical or virtual). The license server requires access to a USB dongle and its license file. The license server is designed for high-load and could provide usage statistics for FlexNet® and OpenIT® monitoring systems.
Geology Designer
Geology Designer allows the user to build a static model from scratch.
Key Features
- Load and edit interpreted seismic surfaces, well trajectories, logs and well picks, facies properties, rock properties, petrophysical information, point sets and other objects. Formats exported from some legacy third-party tools can be loaded to Geology Designer.
- 2D and 3D seismic: import of SEG-Y format in time or depth, creation and visualisation of in-lines, cross-lines and time-slices. Seismic interpretation: time/depth law import. Display of wells, well logs and markers on the Seismic tab (2D).
- Synthetics: calculation to create a synthetic velocity curve, synthetic acoustic impedance, reflection coefficient and interval velocity curves to compare with the seismic data.
- Seismic attributes computation: coherence, instantaneous frequency, amplitude, phase.
- Support for coordinate reference systems by country or EPSG codes.
- 2D and 3D visualisation, histograms, crossplots, vertical proportion curves.
- Well correlation window allows the user to work with many wells at the same time. Automatic and manual well correlation is capable of handling hundreds of wells at the same time. Ability to work with several ghost curves at the same time.
- Faults: loading faults in standard formats, faults creation via polygons, fault editing, build a 3D grid with faults.
- Structural modelling, local grid editing.
- Facies analysis, variograms, property interpolation: least-squares, inverse distance weighing (IDW), kriging, co-kriging, Gaussian simulation (SGS) and multi-point statistic simulation.
- Fluid-in-place in 2D and 3D.
- Calculator to work with all project objects.
- Python based workflows.
- Geosteering.
- All of the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black oil, compositional or thermal), Assisted History Matching and Uncertainty provides the capability to create static and dynamic models in one graphical interface, run simulations, analyse results and carry out assisted history matching and uncertainty analysis.
Combined with the Assisted History Matching & Uncertainty package, the Geology Designer allows the user to capture static uncertainties with full ranges and distributions, offering the ability to carry out a sensitivity and uncertainty analysis on your geological grid, without the need to build a full hydrodynamic model.
Model Designer
Model Designer (pre-processor) allows the user to create a dynamic model and perform local editing, updating and maintenance of the simulation model.
- Loading grid in standard formats, loading RESCUE files.
- Start from existing dynamic model. Property editing, PVT, RP, well production data update.
- Relative permeabilities (Corey and LET correlation, import).
- PVT and EOS: Integration with the PVT Designer.
- VFP tables: Integration with the VFP Designer.
- Rock properties.
- Equilibrium and nonequilibrium initialization.
- Property calculator, local grid editing, aquifers.
- Load and edit well history and events in table form. Pre-defined rules for tables recalculation. Integration with data bases via Python scripts.
- Development Strategy: well groups, limits and control modes, groups limits, economical limits and other rules for wells, well filters.
- Field development planning, restart and forecast scenarios, handling of multiple simulation cases in one Model Designer project.
- 2D and 3D visualization, histograms, crossplots, graphs.
- Python based workflows.
- All the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black-oil, compositional or thermal), Assisted History Matching and Uncertainty provides the capability to create static and dynamic models in one graphical interface, run simulations, analyze results and carry out assisted history matching and uncertainty analysis.
PVT Designer allows the user to create fluid model (PVT, EOS).
- Black oil, compositional and thermal variants.
- Components library, enter user components, calculate component properties via correlations.
- Saturation pressure curve, Phase envelope.
- Hydrates formation and the effect of inhibitors.
- Consideration of non-equilibrium thermodynamic processes.
- Supported lab Experiments: CCE, DLE, CVD, Swelling test, Grading test, Separator test.
- Enter Samples (laboratory data). Run Regression (Matching) – Match points of experiments data (samples) for black oil and compositional models. Set weights for sample points and for experiments independently, Quality check.
- Lumping (create pseudo-components). Use Matching for lumping.
- Splitting of the undefined ‘plus’ fraction.
- Blending compositions together and decontamination (subtracting a known admixture).
- Initialization data. Grading test: composition with respect to depth.
- Create PVT, export PVT tables to create black-oil model.
- EOS. Export EOS data to create compositional model.
- Thermal flash (K-values).
- Create PVT tables via correlations for black-oil cases.
VFP (Well) Designer allows the user to create a well or pipeline model for calculation of lifting tables.
- Create and edit well geometry and construction
- Well geometry. Load a well trajectory in standard formats: Well Path/Deviation, LAS, GWTD, etc. Copy and paste of trajectory points from a text file or a spreadsheet. Manual editing of a trajectory. Well geometry visualization (TVD and Deviation Survey) in 3D.
- Load and visualize LAS in 3D
- Dogleg Severity visualization allows specification of the maximum value of trajectory deviation in degrees per 30 m. A well trajectory is colored depending on its deviation level.
- Well construction specification. Casing, tubing, perforation, squeeze, packer etc. Inflow control devices (ICD): AICD (autonomous) and SICD (spiral). Visualization of devices along well trajectory. Manual drag-and-drop addition and editing of properties of well construction objects. Creation and import of custom object catalogs.
- Selection of parameters of well construction objects as variables for matching experimental data (pressure drop measurements).
- Multilateral well: load and edit trajectories and construction of branches.
- Multisegment well.
- Pressure drop calculations. Different correlation types are available: Beggs & Brill, Hagedorn & Brown, Orkiszewski, Gray and others. Different correlations for vertical, deviated and horizontal parts of the wellbore can be specified. Friction and Hydrostatic component multiplier can be specified
- Calculation of lifting (VFP) tables.
- Normalization of VFP tables.
- Entering experimental data (pressure drop measurements). Visualization of results along with the created VFP tables. Matching of tables by selected measured parameters and variables (network component settings, Friction and Hydrostatic components).
- Creation of IPR table. Available IPR models for gas and liquid: Back pressure, Vogel, Fetkovitch, Jones, Well-PI, Well Test data
- Integration with PVT Designer provides unified fluid properties. PVT models: black-oil, compositional and temperature effects.
- Integration with Network Designer. Provides a well model (well trajectory, construction, VFP tables etc.)
- All the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. Full integration between VFP Designer and other tNavigator modules allows the user to calculate a separate surface network as well as surface coupled with subsurface. Integration with PVT Designer provides unified fluid properties and PVT models: black oil, compositional and temperature effects. Integration with Network Designer provides a well model (well trajectory, construction, VFP tables etc.). In Model Designer each well corresponds to a project of VFP Designer.
Network Designer allows the user to create and calculate the surface network both separately and in a fully coupled way together with the subsurface model. Network Designer is integrated with PVT Designer, Well Designer, Model Designer, Geology Designer, and Simulator.
- Create and edit a surface network. Standard Elements Library is available and contains: well, injector, source, link, pipe, joint, pump, choke, compressor, 2-phase, and 3-phase separators, sink, Python object, gas lift object. Objects that specify upper/lower limits of phase rates and pressure for wells and groups of wells are available: automatic chokes, automatic pumps, limits.
- Steady-state calculation of the surface network to model flow, pressure & temperature given the initial and boundary conditions. Temperature dependence. The effect of pipe burial depth.
- Control of network correctness: detection of network parts, where a flow is absent, before running the calculation; detection of the inconsistency of heights of pipe end-points at pipe joints; control of sufficiency of the number of boundary conditions (pressure, mass flow rate).
- Integration with PVT Designer provides unified fluid properties. PVT models: black oil, compositional and temperature effects, mixing compositional variants (EOS blend).
- Integration with Well Designer provides VFP and IPR tables. Hydrostatic and dynamic pressure losses calculations: different correlations, temperature effects; Burial configuration effects.
- Integration with Model Designer. When importing a model with a network into Model Designer, a Network Designer project will be automatically created and available for further editing. Loading of surface maps and automatic building of pipe profiles with sophisticated geometry.
- Integration with Simulator. A fully implicit coupling surface with subsurface and wells is provided.
- Various tools to analyse and visualise results in a graphical interface are available: Bubble maps, Contribution charts, etc.
- Unified Graphical user interface (GUI) provides a synchronized visualization for integrated models “reservoir+well+surface network”.
- All of the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. At each time step a fully coupled implicit calculation of the integrated model “reservoir+well+surface network” considering the well constraints is provided by the complete integration of Network Designer, PVT Designer, VFP Designer, and Simulator.
Black Oil Simulator typical use cases include simulations of light oil and gas production. It supports all industry standard functionality including live oil, dead oil, and wet gas. This module can be used with the Graphical User Interface (GUI) module or separately as a console version on a workstation or HPC cluster. Input data from some third party data formats are supported.
- General mesh formulation (NNC, LGR and coarsening, faults, pinchouts, etc.), corner point, generalized corner point, unstructured grids.
- Dual porosity, dual permeability.
- 3-phase relative permeabilities with end-point scaling, gravity, capillary effects, saturation, PVT, equilibrium regions.
- API tracking, gravity drainage, nano-polymer flooding, desalination.
- Tracer analysis, waterflood optimization, aquifers, waters with different salinities.
- Polymers, surfactants, ASP injection, BrightWater® polymers.
- Hysteresis, diffusion, adsorption, desorption.
- Extensive support of hydraulic fractures.
- Multisegment wells, group controls, aquifers including constant flux, Fetkovich, Carter-Tracy, numerical. Extended surface network option.
- Temperature extension of black-oil to model the injection of cold or hot water.
- D-factor, GPP controls, VFP lifting tables and correlation functions, ACTIONs, auto-drilling, support of user defined variables, arrays, extended arithmetic.
- Fully implicit and adaptive implicit algorithms.
- Reservoir coupling.
- CPU+GPU processors for faster calculation.
Integration. Integration available with Model Designer, Network Designer and VFP Designer.
The simulation speed-up shown as a function of cores plotted for a workstation with dual physical CPUs.
Compositional Simulator allows the user to simulate compositional models, where PVT properties of oil and gas phases are fitted to an equation of state (EOS), as a mixture of components. Input data from some third party input decks are supported. This module can be used with Graphical Interface module or separately as a console version on a workstation or a cluster.
- Multiple EOS (Peng-Robinson, Redlich-Kwong, Soave-Redlich-Kwong) regions.
- Simulation of non-equilibrium thermodynamic processes.
- Using CPU+GPU processors, clusters for faster calculation.
- CO2 injection, cycling water-gas injection.
- Molecular diffusion, adsorption and desorption, coal bed methane (CBM).
- Relative permeability scaling with respect to composition.
- Special treatment for oil and gas relative permeabilities near the critical point.
- Distribution of CO2 and H2S in water phase.
- Velocity dependent relative permeabilities.
- Gas plants, gas fuel, sales and re-injection, multistage separators.
- Gas Daily Contracted Quantity (DCQ) for gas field model.
- Mixture injection (multicomponent and multiphase – WAG).
- Production and injection surface networks.
- Segments of multi-segment wells that represent sub-critical valves.
- Pressure maintenance regions.
- Reservoir coupling.
Thermal Compositional includes temperature in compositional simulations and is typically used for hot water and steam injection simulations. Input data from some third party thermal decks are supported. This module can be used with Graphical Interface module or separately as a console version on the workstation or cluster.
- K-values for hydrocarbon components via tables or via correlation formulas (surface).
- Four phases: oil (hydrocarbon components), gas (hydrocarbon components, water), water and the solid phase. Phase transitions: evaporation, condensation, dissolution, combustion, modeling of chemical reactions.
- Support for solid phase and chemical reactions for in-situ combustion process.
- Equilibrium and nonequilibrium initialization.
- Porosity dependence on temperature and pressure.
- Liquid phases individual component densities, viscosities as functions of temperature and pressure.
- Enthalpies of hydrocarbon components and rock as functions of temperature.
- Relative permeabilities scaling with respect to composition and temperature.
- Analytical, semi-analytical and numerical aquifers.
- Analytical model of heat exchange with the environment.
- Thermal conductivity dependence on conductivities of mobile phases, solid phases and rock.
- Electrical heaters.
- Dual porosity, dual permeability options.
- Steam injection, mixture injection, multicomponent and multiphase streams, WAG.
- Steam Assisted Gravity Drainage technology (SAGD).
- Using CPU+GPU processors, clusters for faster calculation.
- Reservoir coupling.
Assisted History Matching and Uncertainty analysis
Assisted History Matching (AHM) and Uncertainty Analysis module allows the user to treat any parameter in tNavigator as a variable with a range of uncertainty or as an arithmetic expression. The module includes a Graphical Interface for run control (workstation or cluster) and runtime statistical analysis of the simulation results.
- Experimental design: tornado, Latin hypercube, grid search, Plackett-Burman, Monte-Carlo.
- Optimization algorithms: differential evolution, Single and Multi-objective Particle Swarm optimization (SOPSO and MOPSO), simplex method (Nelder-Mead), response surface (Proxy models can be calculated and exported).
- 3D discrete cosine transform (DCT) algorithm.
- Arbitrary objective function, RFT/MDT incorporation, NPV optimization, UDQ objective function, user-defined functions defined via Python scripts.
- Graphical Interface: graphs, tables, histograms, cross-plots to compare model variants.
- Analytics: Stacked plots, Pareto charts (Pearson and Spearman correlations), multidimensional scaling (MDS), clusterization, table of coefficients R2.
- P10, P50, P90 and other quantiles.
- Forecast optimization, optimization of well position and trajectory.
- Incorporation of workflows from Geology Designer or/and Model Designer.
- Workflows editable in Python scripts.
- Integrated with Job Queue.
- Calculations on workstation or cluster – mouse control of cluster calculation and remote Graphical Interface.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black oil, compositional or thermal), AHM and Uncertainty Analysis provides the possibility to create static and dynamic models in one graphical interface, run simulations, analyze results and carry out a
Graphical User Interface
In tNavigator, run-control monitoring and simulation result postprocessing are done using a single multi-window graphical interface. tNavigator GUI is a universal data analysis and visualization module. It can be used standalone for viewing existing simulation results generated by tNavigator and third party binaries, or integrated with a tNavigator simulation engine to provide interactive run control and instant results monitoring at calculated time steps. The distribution of initial and calculated model grid properties can be viewed as 2D and 3D views, cross-sections, well fences, and 1D or 2D histograms. Calculated and historic production data at the field (when available), group, well, perforation levels can be viewed as graphs, cross-plots, summary tables, bubble maps, contours, and profiles along well trajectories. The interface allows loading of LAS data, trajectories, and comparison with dynamic well profiles. At each time step, the calculated pressure in the grid blocks is used to generate 2D and 3D streamlines. The streamlines are used to calculate injector-producer allocation factors summarized in the form of a drainage table or visualized as 2D drainage network. The 2D map of any grid property can be overlaid by a set of contour lines.
- Mouse control simulation: start, pause, restart.
- Visualization of 2D and 3D dynamic maps and graphs during calculation.
- Graph templates, bubble maps, contours, cross-plots, 1D, 2D histograms, well profiles, reports.
- Graph calculator to create user graphs via Python scripts.
- Waterflood optimization: interactive tracers, streamlines, drainage graphs and colored tables.
- Sector modelling: automatic split and merge.
- Advanced property calculator: build any grid properties and filters to analyze data.
- A Remote Graphical Interface is available to control calculations running on cluster.
In this section, we present a brief description of tNavigator simulation engine.
Equations of fluid flow in porous media. Molar densities and pressure are set to the state variables that allow a general compositional model formulation, where the black oil model is a special case.
The model takes into account the following physical aspects:
- Darcy’s law for fluid flow;
- networks and gas gathering systems
- PVT\\EOS, hysteresis;
- dual porosity, dual permeability;
- streamline, aquifers, tracers;
- vertical, horizontal, multisegment wells, drilling queue;
- coal bed methane model (CBM);
- molecular diffusion, adsorption, desorption;
- enhanced oil recovery (surfactants, polymers, alkaline, foam, CO2 injection);
- steam injection, Steam Assisted Gravity Drainage technology (SAGD);
- and other.
Model grid.
The following aspects are supported:
- block-centered geometry, corner-point geometry, grid specification via blocks tops;
- unstructured grids;
- local grid refinement and coarsening;
- non-neighbor connections, faults, pinchouts;
- multireservoir option;
- reservoir coupling, master-slave;
- sector modeling (automatic split and merge, boundary conditions).
Time approximations.
The following schemes are available to solve systems of differential equations: fully implicit time scheme or AIM (adaptive implicit).
Space approximations.
Finite-volume method with finite-difference approximation of the differential operators is utilized. It assumes upstream approximation.
Non-linear system of equations.
We use the Newton method with full Jacobian with analytic derivatives for solving the non-linear system of equations.
Linear systems with Jacobian and matrix operations.
We use BCGS (BiConjugate Gradient Stabilized) algorithm to solve the system of linear equations. This is a novel method that gets automatically adjusted to the problem at hand. We employ ILU(0) as a preconditioner for the linear system solver. This method is a variation of LU-dcomposition that is designed specially for the tNavigator package.
When solving a linear system with Jacobian one has to store both the sparse matrix and the preconditioner. tNavigator contains a block-oriented modification of the widely used MSR (Modified Sparse Row) method yielding improvement both in memory volume and speed.
Interactive operations and data handling in tNavigator. In order to optimize hard disk operations during the simulation runs, tNavigator treats only state variables of the model. Since the simulation core and the visualizer are in fact one application, the rest of the information can be immediately recovered by the user request. This approach allows operations with the model during the run and quick visualization with no time spent on writing and reading data from the disk.
Supercomputing in Reservoir Simulation
The problem of efficient utilization of computational resources gets more vital taking into account growing demands in the quality and details of 3D reservoir models. Nowadays, reservoir engineers typically have to upscale geological models when dealing with medium and large reservoirs in order to meet the deadlines with the available computation resources. Low resolution details of the dynamic model coupled to the upscalling errors lead to questionable results in production forecasting.
The efficiency of modern computer systems exhibit continuous growth due to increasing numbers of computational cores. High-performance hardware costs are decreasing every day, and hardware-software systems that were extremely expensive just a couple of years ago can be purchased for a reasonable price even for small service companies. Taking into account the availability of multi-CPU computers, we should pay more attention to the software side in order to utilize all computational resources in parallel simulations. Due to outdated software architecture, most common reservoir simulators skip most of the available computation power.
Our technology allows simulations on the finest grids due to utilization of the most recent developments in the software. According to our vision, this approach not only improves the model details, but also assists in collaboration between geologists and reservoir engineers.
tNavigator on workstations
The following novel approaches were implemented in tNavigator in order to increase the efficiency of parallel computations on multi-core workstations:
- All calculations should be carried out in parallel, including linear system solutions, well equations, matrix operations, etc.
- Within each CPU involved in simulation, all data exchange is handled directly by system threads (Boost ~ 30-40%).
- NUMA (Non-Uniform Memory Access) is supported for multi-CPU workstations (Boost ~ 50%).
- Hyperthreading technology support within each CPU (Boost ~15%).
- CPU+GPU technology. Acceleration of solution of system of linear equations using NVidia GPU. The only GPUs starting Pascal architecture and latest CUDA drivers are supported (The speedup from GPU significantly depend on model physics and ratio for CPU to GPU performance).
The basic parallel algorithm in tNavigator is designed for multicore PCs. It is based on direct utilization of system threads, which seems to be optimal for task distribution between different cores within one CPU. This approach allows almost linear acceleration factors when applied on a modern PC.
Now we can see the continuous increase of PC computational efficiency due to progressive growth in the number of cores. Thus, each reservoir engineer can actually get a supercomputer on his desk, and reasonable hardware utilization allows nearly unlimited boost.
Example on the following workstations:
2011: Dual Xeon X5650, (2×6) 12 cores, 2.66GHz, 3 channels DDR3 1333 MHz (e.g. HP Z800)
2012: Dual Xeon E2680, (2×8) 16 cores, 2.7GHz, 4 channels DDR3 1600 MHz (e.g. HP Z820)
2013: Dual Xeon E2697v2, (2×12) 24 cores, 2.7GHz, 4 channels DDR3 1866 MHz (e.g. HP Z820)
2014: Dual Xeon E2697v3, (2×12) 28 cores, 2.6GHz, 4 channels DDR4 2133 MHz (e.g. HP Z840)
2016: Dual Xeon E2699v4, (2×22) 44 cores, 2.2GHz, 4 channels DDR4 2400 MHz (e.g. HP Z840)
tNavigator on clusters
On the workstations the acceleration factor cannot be kept at the same level when the model dimension increases. The problem of memory access speed starts playing a major role in this case. Thus, despite more efficient computers, the simulation time cannot be reduced any further due to memory speed limitations.
This restriction can be removed with distributed memory CPU clusters that require MPI-based algorithms for data interaction between the nodes. This approach is implemented in most of the reservoir simulators. tNavigator contains a novel hybrid algorithm for parallel computations. It utilizes the MPI approach for task distribution between the cluster nodes, and system threads between the cores within each node (MPI+threads). This approach removes the restrictions by utilizing all resources as efficient as possible, and gives a tenfold greater acceleration factor compared to the market leaders on multicore CPU clusters.
In the Moscow office Rock Flow Dynamics utilizes the cluster with 137 nodes (2652 cores) for hydrodynamic simulation. 100 nodes: 2x Xeon E5 2680v2, 20 nodes: 2x Xeon 5650, 8 nodes: 2x Xeon E5 2630v4, 9 nodes: 2x Xeon E5 2680v4, Infiniband 56 Gb/s, RAM 12.8TB, 300TB HDD.
Example of parallel speed-up on clusters.
We are proud to show an example of tNavigator scalability on cluster «Lomonosov» in Moscow State University: 512 nodes, 4096 cores, Xeon X5670, Infiniband QDR 40Gb/s. The three-phase model has been calculated: 20 million active grid blocks, 39 wells. tNavigator delivered the acceleration 1328 times. As we know this is the world record in dynamic simulations. Results have been published in the paper SPE 163090.
Hardware
Joint projects with Intel keep us fairly updated on recent hardware progressive technology. Based on years of research in high performance computing and parallel algorithms in reservoir simulation we gained a high level of expertise with multicore hardware of different types.
We offer our clients utilization of our know-how when looking for optimal software and hardware configurations for reservoir engineering problems. We can recommend or deliver hardware systems based on client needs optimizing the price-quality relationship.
Requirements for RAM per core.
We have no minimal requirements for RAM per core. Everything depends on model. We use 3kB RAM per active grid block for black-oil models.
An example for cluster with the following configuration: Xeon 5650 node, 12 cores, 24Gb.
– One node simulation. We can run 24000k/3k=8million active grid blocks. We have run successfully the real model 6.5 million blocks on cluster with this configuration.
– MPI-version. Simulation on several nodes. For MPI run maximal size of the model multiplies by number of nodes per run. For the cluster with this configuration we can run model with 12 million active grid blocks using 2 nodes and 23 million active with 4 nodes (there is small overhead for domains overlapping in MPI run so maximal size is less than theoretical maximum).
Hardware Recommendations
Workstation
The price range is about $10,000
Example on the following workstations:
2011: Dual Xeon X5650, (2×6) 12 cores, 2.66GHz, 3 channels DDR3 1333 MHz (e.g. HP Z800)
2012: Dual Xeon E2680, (2×8) 16 cores, 2.7GHz, 4 channels DDR3 1600 MHz (e.g. HP Z820)
2013: Dual Xeon E2697v2, (2×12) 24 cores, 2.7GHz, 4 channels DDR3 1866 MHz (e.g. HP Z820)
2014: Dual Xeon E2697v3, (2×12) 28 cores, 2.6GHz, 4 channels DDR4 2133 MHz (e.g. HP Z840)
2016: Dual Xeon E2699v4, (2×22) 44 cores, 2.2GHz, 4 channels DDR4 2400 MHz (e.g. HP Z840)
The standard commercial configuration could be found here: HP Z820 Workstation
Cluster
Here is an example for 8 computing nodes, 160-core configuration.
The price range is about $80,000
Each computing node contains 20-core dual Xeon E5 2680v2 with 2.8 GHz,
128GB DDR3 1866GHz RAM, Infiniband network switch with 56Gb/s
Typical parallel acceleration:
Model 1. 2.4 million active grid blocks, single porosity, 3-phase blackoil model, gas cap, and multiple equilibrium regions, 39 wells. The parallel acceleration factor is 73 times
Model 2. 460 thousand active grid blocks, dual porosity, 3 phase blackoil model with 4 horizontal wells(~3km each) and multiple hydraulic fractures represented by 250 Local
Grid Refinements(LGRs). It is an example of shale oil reservoir. The parallel acceleration factor is 40 times
Model 3. 1.9 million active grid block, compositional EOS model, with 6 components and CO2 injection, 4 horizontal wells(~3km long). The parallel acceleration factor is 80 times
Geology Designer
Geology Designer allows the user to build a static model from scratch.
Key Features
- Load and edit interpreted seismic surfaces, well trajectories, logs and well picks, facies properties, rock properties, petrophysical information, point sets and other objects. Formats exported from some legacy third-party tools can be loaded to Geology Designer.
- 2D and 3D seismic: import of SEG-Y format in time or depth, creation and visualisation of in-lines, cross-lines and time-slices. Seismic interpretation: time/depth law import. Display of wells, well logs and markers on the Seismic tab (2D).
- Synthetics: calculation to create a synthetic velocity curve, synthetic acoustic impedance, reflection coefficient and interval velocity curves to compare with the seismic data.
- Seismic attributes computation: coherence, instantaneous frequency, amplitude, phase.
- Support for coordinate reference systems by country or EPSG codes.
- 2D and 3D visualisation, histograms, crossplots, vertical proportion curves.
- Well correlation window allows the user to work with many wells at the same time. Automatic and manual well correlation is capable of handling hundreds of wells at the same time. Ability to work with several ghost curves at the same time.
- Faults: loading faults in standard formats, faults creation via polygons, fault editing, build a 3D grid with faults.
- Structural modelling, local grid editing.
- Facies analysis, variograms, property interpolation: least-squares, inverse distance weighing (IDW), kriging, co-kriging, Gaussian simulation (SGS) and multi-point statistic simulation.
- Fluid-in-place in 2D and 3D.
- Calculator to work with all project objects.
- Python based workflows.
- Geosteering.
- All of the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black oil, compositional or thermal), Assisted History Matching and Uncertainty provides the capability to create static and dynamic models in one graphical interface, run simulations, analyse results and carry out assisted history matching and uncertainty analysis.
Combined with the Assisted History Matching & Uncertainty package, the Geology Designer allows the user to capture static uncertainties with full ranges and distributions, offering the ability to carry out a sensitivity and uncertainty analysis on your geological grid, without the need to build a full hydrodynamic model.
Model Designer
Model Designer (pre-processor) allows the user to create a dynamic model and perform local editing, updating and maintenance of the simulation model.
- Loading grid in standard formats, loading RESCUE files.
- Start from existing dynamic model. Property editing, PVT, RP, well production data update.
- Relative permeabilities (Corey and LET correlation, import).
- PVT and EOS: Integration with the PVT Designer.
- VFP tables: Integration with the VFP Designer.
- Rock properties.
- Equilibrium and nonequilibrium initialization.
- Property calculator, local grid editing, aquifers.
- Load and edit well history and events in table form. Pre-defined rules for tables recalculation. Integration with data bases via Python scripts.
- Development Strategy: well groups, limits and control modes, groups limits, economical limits and other rules for wells, well filters.
- Field development planning, restart and forecast scenarios, handling of multiple simulation cases in one Model Designer project.
- 2D and 3D visualization, histograms, crossplots, graphs.
- Python based workflows.
- All the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black-oil, compositional or thermal), Assisted History Matching and Uncertainty provides the capability to create static and dynamic models in one graphical interface, run simulations, analyze results and carry out assisted history matching and uncertainty analysis.
Model Designer for Unconventional Reservoirs
Model Designer enables users to design and simulate multi-stage hydraulic fracture configuration with virtually unlimited complexity.
- Fractures with arbitrary geometry can be modelled with no restrictions on the number of fracture stages and fracture clusters; nor on the angles of fracture-to-fracture, fracture-to-well, fracture-to-grid relative orientations.
- Various rock properties (porosity, permeability, net-to-gross ratio) and reservoir regions (saturation, rock compaction, PVT) can be initialized and changed in time via schedule section separately for fracture and non-fracture zones within stimulated rock volume (SRV).
- Fractures and SRV zones are defined through one or several Templates with parameters included in the Assisted History Matching and Uncertainty Quantification workflow.
- Multiple fractures can merge and split from each other. A new type of adaptive logarithmic LGRs can be used to ensure effective unstructured gridding around fracture paths.
- Fracture properties for multiple wells can be input into the project via Fracture Table that enables handling large data arrays.
- Incorporation of Frac Propagation results from third party software (GOHFER®, StimPlan™, etc.).
- Dual porosity/Dual Permeability, Coal Bed Methane and other options are available.
- Assessment of the fracture parameters’ impact on production (uncertainty analysis via workflow).
PVT Designer
PVT Designer allows the user to create fluid model (PVT, EOS).
- Black oil, compositional and thermal variants.
- Components library, enter user components, calculate component properties via correlations.
- Saturation pressure curve, Phase envelope.
- Hydrates formation and the effect of inhibitors.
- Consideration of non-equilibrium thermodynamic processes.
- Supported lab Experiments: CCE, DLE, CVD, Swelling test, Grading test, Separator test.
- Enter Samples (laboratory data). Run Regression (Matching) – Match points of experiments data (samples) for black oil and compositional models. Set weights for sample points and for experiments independently, Quality check.
- Lumping (create pseudo-components). Use Matching for lumping.
- Splitting of the undefined ‘plus’ fraction.
- Blending compositions together and decontamination (subtracting a known admixture).
- Initialization data. Grading test: composition with respect to depth.
- Create PVT, export PVT tables to create black-oil model.
- EOS. Export EOS data to create compositional model.
- Thermal flash (K-values).
- Create PVT tables via correlations for black-oil cases.
VFP Designer
VFP (Well) Designer allows the user to create a well or pipeline model for calculation of lifting tables.
- Create and edit well geometry and construction
- Well geometry. Load a well trajectory in standard formats: Well Path/Deviation, LAS, GWTD, etc. Copy and paste of trajectory points from a text file or a spreadsheet. Manual editing of a trajectory. Well geometry visualization (TVD and Deviation Survey) in 3D.
- Load and visualize LAS in 3D
- Dogleg Severity visualization allows specification of the maximum value of trajectory deviation in degrees per 30 m. A well trajectory is colored depending on its deviation level.
- Well construction specification. Casing, tubing, perforation, squeeze, packer etc. Inflow control devices (ICD): AICD (autonomous) and SICD (spiral). Visualization of devices along well trajectory. Manual drag-and-drop addition and editing of properties of well construction objects. Creation and import of custom object catalogs.
- Selection of parameters of well construction objects as variables for matching experimental data (pressure drop measurements).
- Multilateral well: load and edit trajectories and construction of branches.
- Multisegment well.
- Pressure drop calculations. Different correlation types are available: Beggs & Brill, Hagedorn & Brown, Orkiszewski, Gray and others. Different correlations for vertical, deviated and horizontal parts of the wellbore can be specified. Friction and Hydrostatic component multiplier can be specified
- Calculation of lifting (VFP) tables.
- Normalization of VFP tables.
- Entering experimental data (pressure drop measurements). Visualization of results along with the created VFP tables. Matching of tables by selected measured parameters and variables (network component settings, Friction and Hydrostatic components).
- Creation of IPR table. Available IPR models for gas and liquid: Back pressure, Vogel, Fetkovitch, Jones, Well-PI, Well Test data
- Integration with PVT Designer provides unified fluid properties. PVT models: black-oil, compositional and temperature effects.
- Integration with Network Designer. Provides a well model (well trajectory, construction, VFP tables etc.)
- All the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. Full integration between VFP Designer and other tNavigator modules allows the user to calculate a separate surface network as well as surface coupled with subsurface. Integration with PVT Designer provides unified fluid properties and PVT models: black oil, compositional and temperature effects. Integration with Network Designer provides a well model (well trajectory, construction, VFP tables etc.). In Model Designer each well corresponds to a project of VFP Designer.
Network Designer
Network Designer allows the user to create and calculate the surface network both separately and in a fully coupled way together with the subsurface model. Network Designer is integrated with PVT Designer, Well Designer, Model Designer, Geology Designer, and Simulator.
- Create and edit a surface network. Standard Elements Library is available and contains: well, injector, source, link, pipe, joint, pump, choke, compressor, 2-phase, and 3-phase separators, sink, Python object, gas lift object. Objects that specify upper/lower limits of phase rates and pressure for wells and groups of wells are available: automatic chokes, automatic pumps, limits.
- Steady-state calculation of the surface network to model flow, pressure & temperature given the initial and boundary conditions. Temperature dependence. The effect of pipe burial depth.
- Control of network correctness: detection of network parts, where a flow is absent, before running the calculation; detection of the inconsistency of heights of pipe end-points at pipe joints; control of sufficiency of the number of boundary conditions (pressure, mass flow rate).
- Integration with PVT Designer provides unified fluid properties. PVT models: black oil, compositional and temperature effects, mixing compositional variants (EOS blend).
- Integration with Well Designer provides VFP and IPR tables. Hydrostatic and dynamic pressure losses calculations: different correlations, temperature effects; Burial configuration effects.
- Integration with Model Designer. When importing a model with a network into Model Designer, a Network Designer project will be automatically created and available for further editing. Loading of surface maps and automatic building of pipe profiles with sophisticated geometry.
- Integration with Simulator. A fully implicit coupling surface with subsurface and wells is provided.
- Various tools to analyse and visualise results in a graphical interface are available: Bubble maps, Contribution charts, etc.
- Unified Graphical user interface (GUI) provides a synchronized visualization for integrated models “reservoir+well+surface network”.
- All of the calculations are accelerated with parallel algorithms run on all available cores of the hardware.
Integration. At each time step a fully coupled implicit calculation of the integrated model “reservoir+well+surface network” considering the well constraints is provided by the complete integration of Network Designer, PVT Designer, VFP Designer, and Simulator.
Black Oil Simulator
Black Oil Simulator typical use cases include simulations of light oil and gas production. It supports all industry standard functionality including live oil, dead oil, and wet gas. This module can be used with the Graphical User Interface (GUI) module or separately as a console version on a workstation or HPC cluster. Input data from some third party data formats are supported.
- General mesh formulation (NNC, LGR and coarsening, faults, pinchouts, etc.), corner point, generalized corner point, unstructured grids.
- Dual porosity, dual permeability.
- 3-phase relative permeabilities with end-point scaling, gravity, capillary effects, saturation, PVT, equilibrium regions.
- API tracking, gravity drainage, nano-polymer flooding, desalination.
- Tracer analysis, waterflood optimization, aquifers, waters with different salinities.
- Polymers, surfactants, ASP injection, BrightWater® polymers.
- Hysteresis, diffusion, adsorption, desorption.
- Extensive support of hydraulic fractures.
- Multisegment wells, group controls, aquifers including constant flux, Fetkovich, Carter-Tracy, numerical. Extended surface network option.
- Temperature extension of black-oil to model the injection of cold or hot water.
- D-factor, GPP controls, VFP lifting tables and correlation functions, ACTIONs, auto-drilling, support of user defined variables, arrays, extended arithmetic.
- Fully implicit and adaptive implicit algorithms.
- Reservoir coupling.
- CPU+GPU processors for faster calculation.
Integration. Integration available with Model Designer, Network Designer and VFP Designer.
Compositional Simulator
Compositional Simulator allows the user to simulate compositional models, where PVT properties of oil and gas phases are fitted to an equation of state (EOS), as a mixture of components. Input data from some third party input decks are supported. This module can be used with Graphical Interface module or separately as a console version on a workstation or a cluster.
- Multiple EOS (Peng-Robinson, Redlich-Kwong, Soave-Redlich-Kwong) regions.
- Simulation of non-equilibrium thermodynamic processes.
- Using CPU+GPU processors, clusters for faster calculation.
- CO2 injection, cycling water-gas injection.
- Molecular diffusion, adsorption and desorption, coal bed methane (CBM).
- Relative permeability scaling with respect to composition.
- Special treatment for oil and gas relative permeabilities near the critical point.
- Distribution of CO2 and H2S in water phase.
- Velocity dependent relative permeabilities.
- Gas plants, gas fuel, sales and re-injection, multistage separators.
- Gas Daily Contracted Quantity (DCQ) for gas field model.
- Mixture injection (multicomponent and multiphase – WAG).
- Production and injection surface networks.
- Segments of multi-segment wells that represent sub-critical valves.
- Pressure maintenance regions.
- Reservoir coupling.
Thermal Compositional Simulator
Thermal Compositional includes temperature in compositional simulations and is typically used for hot water and steam injection simulations. Input data from some third party thermal decks are supported. This module can be used with Graphical Interface module or separately as a console version on the workstation or cluster.
- K-values for hydrocarbon components via tables or via correlation formulas (surface).
- Four phases: oil (hydrocarbon components), gas (hydrocarbon components, water), water and the solid phase. Phase transitions: evaporation, condensation, dissolution, combustion, modeling of chemical reactions.
- Support for solid phase and chemical reactions for in-situ combustion process.
- Equilibrium and nonequilibrium initialization.
- Porosity dependence on temperature and pressure.
- Liquid phases individual component densities, viscosities as functions of temperature and pressure.
- Enthalpies of hydrocarbon components and rock as functions of temperature.
- Relative permeabilities scaling with respect to composition and temperature.
- Analytical, semi-analytical and numerical aquifers.
- Analytical model of heat exchange with the environment.
- Thermal conductivity dependence on conductivities of mobile phases, solid phases and rock.
- Electrical heaters.
- Dual porosity, dual permeability options.
- Steam injection, mixture injection, multicomponent and multiphase streams, WAG.
- Steam Assisted Gravity Drainage technology (SAGD).
- Using CPU+GPU processors, clusters for faster calculation.
- Reservoir coupling.
Fully Coupled Geomechanics Simulations
tNavigator uses a joint system of coupled equations to describe filtration processes in the reservoir and geomechanical effects on the unified grid.
- The same model grid is used for reservoir dynamic and geomechanic simulations (block centers for reservoir simulations, block corners for geomechanics).
- Joint system of coupled equations is solved numerically in a parallel way: geomechanics (only CPU cores), reservoir dynamics (CPU and GPU cores).
- All model types (black oil, thermal, compositional) are supported.
- Young’s modulus, Poisson constant, boundary condition for stress, boundary condition for displacement.
- Modeling of geomechanical effects via hysteresis rock compaction data tables.
- Mohr-Coulomb failure criterion is used for analysis of stress state and predict the potential rock fault. Possible fracture directions.
Assisted History Matching and Uncertainty analysis
Assisted History Matching (AHM) and Uncertainty Analysis module allows the user to treat any parameter in tNavigator as a variable with a range of uncertainty or as an arithmetic expression. The module includes a Graphical Interface for run control (workstation or cluster) and runtime statistical analysis of the simulation results.
- Experimental design: tornado, Latin hypercube, grid search, Plackett-Burman, Monte-Carlo.
- Optimization algorithms: differential evolution, Single and Multi-objective Particle Swarm optimization (SOPSO and MOPSO), simplex method (Nelder-Mead), response surface (Proxy models can be calculated and exported).
- 3D discrete cosine transform (DCT) algorithm.
- Arbitrary objective function, RFT/MDT incorporation, NPV optimization, UDQ objective function, user-defined functions defined via Python scripts.
- Graphical Interface: graphs, tables, histograms, cross-plots to compare model variants.
- Analytics: Stacked plots, Pareto charts (Pearson and Spearman correlations), multidimensional scaling (MDS), clusterization, table of coefficients R2.
- P10, P50, P90 and other quantiles.
- Forecast optimization, optimization of well position and trajectory.
- Incorporation of workflows from Geology Designer or/and Model Designer.
- Workflows editable in Python scripts.
- Integrated with Job Queue.
- Calculations on workstation or cluster – mouse control of cluster calculation and remote Graphical Interface.
Integration. The combination of modules Geology Designer, Model Designer, PVT Designer, VFP Designer, Simulator (black oil, compositional or thermal), AHM and Uncertainty Analysis provides the possibility to create static and dynamic models in one graphical interface, run simulations, analyze results and carry out assisted history matching and uncertainty analysis.
Fully Integrated Modelling
Assisted History Matching module provides comprehensive sensitivity analysis of simulation results with respect to variations of static and dynamic parameters defined by workflow. The workflow may include step-by step building of a structural model in Geology Designer followed by snapping seismic surfaces to match markers, grid generation, upscaling, SGS property interpolation and dynamic model initialization with static and dynamic uncertainty variables.
Graphical User Interface
In tNavigator, run-control monitoring and simulation result postprocessing are done using a single multi-window graphical interface. tNavigator GUI is a universal data analysis and visualization module. It can be used standalone for viewing existing simulation results generated by tNavigator and third party binaries, or integrated with a tNavigator simulation engine to provide interactive run control and instant results monitoring at calculated time steps. The distribution of initial and calculated model grid properties can be viewed as 2D and 3D views, cross-sections, well fences, and 1D or 2D histograms. Calculated and historic production data at the field (when available), group, well, perforation levels can be viewed as graphs, cross-plots, summary tables, bubble maps, contours, and profiles along well trajectories. The interface allows loading of LAS data, trajectories, and comparison with dynamic well profiles. At each time step, the calculated pressure in the grid blocks is used to generate 2D and 3D streamlines. The streamlines are used to calculate injector-producer allocation factors summarized in the form of a drainage table or visualized as 2D drainage network. The 2D map of any grid property can be overlaid by a set of contour lines.
- Mouse control simulation: start, pause, restart.
- Visualization of 2D and 3D dynamic maps and graphs during calculation.
- Graph templates, bubble maps, contours, cross-plots, 1D, 2D histograms, well profiles, reports.
- Graph calculator to create user graphs via Python scripts.
- Waterflood optimization: interactive tracers, streamlines, drainage graphs and colored tables.
- Sector modelling: automatic split and merge.
- Advanced property calculator: build any grid properties and filters to analyze data.
- A Remote Graphical Interface is available to control calculations running on cluster.
tNavigator Technical Description
In this section, we present a brief description of tNavigator simulation engine.
Equations of fluid flow in porous media. Molar densities and pressure are set to the state variables that allow a general compositional model formulation, where the black oil model is a special case.
The model takes into account the following physical aspects:
- Darcy’s law for fluid flow;
- networks and gas gathering systems
- PVT\\EOS, hysteresis;
- dual porosity, dual permeability;
- streamline, aquifers, tracers;
- vertical, horizontal, multisegment wells, drilling queue;
- coal bed methane model (CBM);
- molecular diffusion, adsorption, desorption;
- enhanced oil recovery (surfactants, polymers, alkaline, foam, CO2 injection);
- steam injection, Steam Assisted Gravity Drainage technology (SAGD);
- and other.
Model grid.
The following aspects are supported:
- block-centered geometry, corner-point geometry, grid specification via blocks tops;
- unstructured grids;
- local grid refinement and coarsening;
- non-neighbor connections, faults, pinchouts;
- multireservoir option;
- reservoir coupling, master-slave;
- sector modeling (automatic split and merge, boundary conditions).
Time approximations.
The following schemes are available to solve systems of differential equations: fully implicit time scheme or AIM (adaptive implicit).
Space approximations.
Finite-volume method with finite-difference approximation of the differential operators is utilized. It assumes upstream approximation.
Non-linear system of equations.
We use the Newton method with full Jacobian with analytic derivatives for solving the non-linear system of equations.
Linear systems with Jacobian and matrix operations.
We use BCGS (BiConjugate Gradient Stabilized) algorithm to solve the system of linear equations. This is a novel method that gets automatically adjusted to the problem at hand. We employ ILU(0) as a preconditioner for the linear system solver. This method is a variation of LU-dcomposition that is designed specially for the tNavigator package.
When solving a linear system with Jacobian one has to store both the sparse matrix and the preconditioner. tNavigator contains a block-oriented modification of the widely used MSR (Modified Sparse Row) method yielding improvement both in memory volume and speed.
Interactive operations and data handling in tNavigator.
In order to optimize hard disk operations during the simulation runs, tNavigator treats only state variables of the model. Since the simulation core and the visualizer are in fact one application, the rest of the information can be immediately recovered by the user request. This approach allows operations with the model during the run and quick visualization with no time spent on writing and reading data from the disk.
Supercomputing in Reservoir Simulation
The problem of efficient utilization of computational resources gets more vital taking into account growing demands in the quality and details of 3D reservoir models. Nowadays, reservoir engineers typically have to upscale geological models when dealing with medium and large reservoirs in order to meet the deadlines with the available computation resources. Low resolution details of the dynamic model coupled to the upscalling errors lead to questionable results in production forecasting.
The efficiency of modern computer systems exhibit continuous growth due to increasing numbers of computational cores. High-performance hardware costs are decreasing every day, and hardware-software systems that were extremely expensive just a couple of years ago can be purchased for a reasonable price even for small service companies. Taking into account the availability of multi-CPU computers, we should pay more attention to the software side in order to utilize all computational resources in parallel simulations. Due to outdated software architecture, most common reservoir simulators skip most of the available computation power.
Our technology allows simulations on the finest grids due to utilization of the most recent developments in the software. According to our vision, this approach not only improves the model details, but also assists in collaboration between geologists and reservoir engineers.
tNavigator on workstations
The following novel approaches were implemented in tNavigator in order to increase the efficiency of parallel computations on multi-core workstations:
- All calculations should be carried out in parallel, including linear system solutions, well equations, matrix operations, etc.
- Within each CPU involved in simulation, all data exchange is handled directly by system threads (Boost ~ 30-40%).
- NUMA (Non-Uniform Memory Access) is supported for multi-CPU workstations (Boost ~ 50%).
- Hyperthreading technology support within each CPU (Boost ~15%).
- CPU+GPU technology. Acceleration of solution of system of linear equations using NVidia GPU. The only GPUs starting Pascal architecture and latest CUDA drivers are supported (The speedup from GPU significantly depend on model physics and ratio for CPU to GPU performance).
The basic parallel algorithm in tNavigator is designed for multicore PCs. It is based on direct utilization of system threads, which seems to be optimal for task distribution between different cores within one CPU. This approach allows almost linear acceleration factors when applied on a modern PC.
Now we can see the continuous increase of PC computational efficiency due to progressive growth in the number of cores. Thus, each reservoir engineer can actually get a supercomputer on his desk, and reasonable hardware utilization allows nearly unlimited boost.
Example on the following workstations:
2011: Dual Xeon X5650, (2×6) 12 cores, 2.66GHz, 3 channels DDR3 1333 MHz (e.g. HP Z800)
2012: Dual Xeon E2680, (2×8) 16 cores, 2.7GHz, 4 channels DDR3 1600 MHz (e.g. HP Z820)
2013: Dual Xeon E2697v2, (2×12) 24 cores, 2.7GHz, 4 channels DDR3 1866 MHz (e.g. HP Z820)
2014: Dual Xeon E2697v3, (2×12) 28 cores, 2.6GHz, 4 channels DDR4 2133 MHz (e.g. HP Z840)
2016: Dual Xeon E2699v4, (2×22) 44 cores, 2.2GHz, 4 channels DDR4 2400 MHz (e.g. HP Z840)
tNavigator on clusters
On the workstations the acceleration factor cannot be kept at the same level when the model dimension increases. The problem of memory access speed starts playing a major role in this case. Thus, despite more efficient computers, the simulation time cannot be reduced any further due to memory speed limitations.
This restriction can be removed with distributed memory CPU clusters that require MPI-based algorithms for data interaction between the nodes. This approach is implemented in most of the reservoir simulators. tNavigator contains a novel hybrid algorithm for parallel computations. It utilizes the MPI approach for task distribution between the cluster nodes, and system threads between the cores within each node (MPI+threads). This approach removes the restrictions by utilizing all resources as efficient as possible, and gives a tenfold greater acceleration factor compared to the market leaders on multicore CPU clusters.
In the Moscow office Rock Flow Dynamics utilizes the cluster with 137 nodes (2652 cores) for hydrodynamic simulation. 100 nodes: 2x Xeon E5 2680v2, 20 nodes: 2x Xeon 5650, 8 nodes: 2x Xeon E5 2630v4, 9 nodes: 2x Xeon E5 2680v4, Infiniband 56 Gb/s, RAM 12.8TB, 300TB HDD.
Example of parallel speed-up on clusters.
We are proud to show an example of tNavigator scalability on cluster «Lomonosov» in Moscow State University: 512 nodes, 4096 cores, Xeon X5670, Infiniband QDR 40Gb/s. The three-phase model has been calculated: 20 million active grid blocks, 39 wells. tNavigator delivered the acceleration 1328 times. As we know this is the world record in dynamic simulations. Results have been published in the paper SPE 163090.
We are actively working on developing more efficient parallel computations. We believe there are higher limits we can achieve.
Hardware
Joint projects with Intel keep us fairly updated on recent hardware progressive technology. Based on years of research in high performance computing and parallel algorithms in reservoir simulation we gained a high level of expertise with multicore hardware of different types.
We offer our clients utilization of our know-how when looking for optimal software and hardware configurations for reservoir engineering problems. We can recommend or deliver hardware systems based on client needs optimizing the price-quality relationship.
Requirements for RAM per core.
We have no minimal requirements for RAM per core. Everything depends on model. We use 3kB RAM per active grid block for black-oil models.
An example for cluster with the following configuration: Xeon 5650 node, 12 cores, 24Gb.
– One node simulation. We can run 24000k/3k=8million active grid blocks. We have run successfully the real model 6.5 million blocks on cluster with this configuration.
– MPI-version. Simulation on several nodes. For MPI run maximal size of the model multiplies by number of nodes per run. For the cluster with this configuration we can run model with 12 million active grid blocks using 2 nodes and 23 million active with 4 nodes (there is small overhead for domains overlapping in MPI run so maximal size is less than theoretical maximum).
Hardware Recommendations
Workstation
The price range is about $10,000
Example on the following workstations:
2011: Dual Xeon X5650, (2×6) 12 cores, 2.66GHz, 3 channels DDR3 1333 MHz (e.g. HP Z800)
2012: Dual Xeon E2680, (2×8) 16 cores, 2.7GHz, 4 channels DDR3 1600 MHz (e.g. HP Z820)
2013: Dual Xeon E2697v2, (2×12) 24 cores, 2.7GHz, 4 channels DDR3 1866 MHz (e.g. HP Z820)
2014: Dual Xeon E2697v3, (2×12) 28 cores, 2.6GHz, 4 channels DDR4 2133 MHz (e.g. HP Z840)
2016: Dual Xeon E2699v4, (2×22) 44 cores, 2.2GHz, 4 channels DDR4 2400 MHz (e.g. HP Z840)
The standard commercial configuration could be found here: HP Z820 Workstation
Cluster
Here is an example for 8 computing nodes, 160-core configuration.
The price range is about $80,000
Each computing node contains 20-core dual Xeon E5 2680v2 with 2.8 GHz,
128GB DDR3 1866GHz RAM, Infiniband network switch with 56Gb/s
Typical parallel acceleration:
Model 1. 2.4 million active grid blocks, single porosity, 3-phase blackoil model, gas cap, and multiple equilibrium regions, 39 wells. The parallel acceleration factor is 73 times
Model 2. 460 thousand active grid blocks, dual porosity, 3 phase blackoil model with 4 horizontal wells(~3km each) and multiple hydraulic fractures represented by 250 Local
Grid Refinements(LGRs). It is an example of shale oil reservoir. The parallel acceleration factor is 40 times
Model 3. 1.9 million active grid block, compositional EOS model, with 6 components and CO2 injection, 4 horizontal wells(~3km long). The parallel acceleration factor is 80 times