CONVERGE is a leading computational fluid dynamics (CFD) software package with an emphasis on accuracy, efficiency, and innovation. With truly autonomous meshing, state-of-the-art physical models, and the ability to easily accommodate complex moving geometries, CONVERGE is fully equipped to help you solve the hard problems.
Innovative Meshing
CONVERGE features fully autonomous meshing, which eliminates all user meshing time from the simulation process. In addition, CONVERGE’s novel cut-cell approach perfectly represents your geometry—no matter how complex—and easily accommodates moving boundaries. This approach avoids the numerical viscosity generated by deforming meshes and offers accurate results without intensive hands-on setup. Moreover, Adaptive Mesh Refinement allows you to efficiently achieve the best solution possible for a given computational expense by adding cells when and where they are needed to resolve key flow phenomena.
Capture Essential Physical Processes
Going beyond a visually appealing simulation to obtaining useful, realistic results requires accurate physical modeling. CONVERGE contains an extensive suite of well-validated physical models for simulating everything from fluid-structure interaction and conjugate heat transfer to spray and combustion. In addition, CONVERGE includes the SAGE detailed chemistry solver, which is fully coupled with the flow solver for maximum accuracy and efficiency. If you’re interested in implementing a custom model, you can easily do so through user defined functions (UDFs) to customize CONVERGE to meet your needs.
Accelerate R&D
CONVERGE is designed to simplify and expedite the research and design process for an expansive range of applications, from gas turbine engines to mechanical heart valves. With CONVERGE, you can perform a comprehensive system analysis and optimization to find the best design before building an expensive physical prototype. Since 3D CFD simulations can require long runtimes, CONVERGE enables highly parallel simulations on many processors and demonstrates excellent scaling even on thousands of cores. Taking advantage of these capabilities can drastically reduce time-to-solution for your simulations. Overall, incorporating CONVERGE into your R&D workflow can reduce costs across the board and enable you to bring your product to market sooner.
What’s New in CONVERGE 3.1?
CONVERGE 3.1, the latest major release of our CFD software, includes many exciting new features and enhancements that expand both the capability and usability of the code.
Emobility Modeling Enhancements
As the number of electric vehicles on the road rises, designing optimized battery packs and electric motors is increasingly important. Version 3.1 includes a number of new models and features that augment CONVERGE’s emobility modeling capabilities.
Battery Modeling
CONVERGE 3.1 offers two methods for modeling battery heat sources: the equivalent circuit model and the electric potential solver. With the equivalent circuit model, you can represent your battery as an electrical network and calculate the heat generation based on the current. This model can be used for either charging or discharging batteries. The electric potential solver, useful for batteries and other direct current applications, predicts the current and associated heat transfer based on the electric potential and electrical conductivity of the solid.
For simulating battery thermal runaway, CONVERGE now includes the Hatchard-Kim and Ren mechanisms. In addition, you can easily implement any thermal runaway mechanism via user defined function.
JMAG Coupling for Motor Design
CONVERGE 3.1 adds the capability to couple with JMAG-Designer, JSOL Corporation’s leading electromagnetic solver for electric motor design. JMAG can provide CONVERGE with realistic distributions of heat generation due to electromagnetic losses for use in motor cooling simulations. In CONVERGE Studio, you can directly import JMAG NASTRAN files. CONVERGE Studio will read the geometry and automatically assign thermal boundary conditions or a volumetric heat generation source for conjugate heat transfer (CHT) simulations.
After CONVERGE performs the flow simulation, you can export the resulting heat transfer coefficients to JMAG, where they can be used for another round of electromagnetic analysis. With this coupled methodology, you can accurately simulate a variety of cooling methods, including air cooling, oil cooling, and water jacket cooling.
Implicit Fluid-Structure Interaction Modeling
Fluid-structure interaction (FSI) simulations can become numerically unstable when the density of a fluid and a submerged solid object are similar, or when a solid body is floating on a free surface. CONVERGE 3.1 introduces implicit FSI modeling, which overcomes this limitation. Implicit FSI implements an inner coupling loop between the CFD solver and the FSI solver to more accurately account for the effect of the fluid on the FSI object. With implicit FSI, CONVERGE can accurately simulate applications including floating platforms for offshore wind turbines, boat and ship hulls, and subsea oil and gas applications.
Wind and Wave Simulations
For many of the offshore and marine applications enabled by implicit FSI, being able to generate realistic wind and wave fields is critical to achieving a simulation reflective of the real world. In CONVERGE, the wave generation and synthetic turbulence generation tools allow you to easily simulate regular or irregular 3D waves and introduce realistic turbulence into the wind and wave fields. Since the wind affects the dynamics of the waves and vice versa, CONVERGE’s robust volume of fluid modeling, complete with several interface capturing schemes, can resolve the wind-wave interactions.
Mooring Model
Floating platforms for offshore wind turbines and oil and gas applications are typically constrained by mooring cables. The mooring model in CONVERGE 3.1 employs a finite segment method to efficiently calculate the applied forces from the mooring cables, which can then be used in an FSI simulation.
Volume of Fluid Modeling Enhancements
CONVERGE 3.1 adds several new volume of fluid (VOF) modeling options that enhance the existing multi-phase modeling capabilities.
The flux-corrected transport (FCT) scheme limits both diffusive and dispersive numerical errors to help maintain a sharp interface between fluids. FCT can be used for simulations with any number of gas or liquid species, and it can be used for incompressible or compressible simulations. Coupling the FCT method with Adaptive Mesh Refinement helps to further reduce numerical diffusion.
The newly added mixing model can be used to simulate the separation of phases, or the separation of immiscible liquids, due to gravity by applying independent drift velocities to the fluids. In addition, the mixture model enables the simulation of different flow regimes (e.g., bubbly, slurry), droplet- and particle-laden flows, pneumatic transport, etc. For simulations with larger timesteps, the mixture model helps counteract numerical diffusion to keep the fluids separate. This approach makes it possible to run faster VOF simulations with less of an accuracy loss.
Surface compression is a surface sharpening technique that reduces numerical diffusion at fluid interfaces. This approach is particularly useful for achieving visually realistic results as it enables you to capture smaller scale structures due to the lessened numerical diffusion.
Multi-Stream Simulations
With demand for multi-physics simulation tools rising, CONVERGE 3.1 now offers a multi-stream simulation capability. In CONVERGE, a stream is a sub-domain of the simulation defined by a group of regions. This new capability enables you to apply different physical models and solver settings (e.g., combustion models, turbulence models, reaction mechanisms, time-steps, inputs and outputs, etc.) to different streams. Each stream can work independently or they can be coupled together. For example, consider a simulation of an engine cylinder with an oil gallery for piston cooling. With the new multi-stream capability, you can now have gas flow and combustion occurring in the engine cylinder, oil and air VOF modeling in the oil gallery, and heat transfer within the piston all simulated together as coupled streams. The multi-stream capability makes complex multi-physics simulations possible with a significantly simplified workflow.
In Situ Post-Processing with ParaView Catalyst
During a typical CFD simulation, the solver writes results to a disk at a predefined interval. However, for large cases or cases in which you want frequent data points, writing the data to a disk can take a significant amount of time and storage. With CONVERGE 3.1, you can couple CONVERGE with ParaView Catalyst to enable in situ post-processing of your CFD simulations. With in situ post-processing, CONVERGE sends data to ParaView Catalyst at a user-specified interval, and ParaView performs the post-processing directly on the cluster. Not only does this reduce the required storage space, but it also eliminates all user time associated with post-processing the results at the end of the simulation.
In situ post-processing is set up in CONVERGE Studio. You can choose from simple post-processing operations included in CONVERGE Studio—for example, extracting and saving slices or iso-volumes—or you can import a custom script for advanced post-processing procedures. Depending on the size of your case and your post-processing workflow, in situ post-processing can save you a significant amount of time. In addition, this feature makes it easy to implement your own fully automated post-processing procedure via custom scripts.
Moving Boundaries Update
In previous versions of CONVERGE, there was a limitation as to how far a wall could move during a single time-step. In CONVERGE 3.1, this restriction has been removed—walls can now move through an arbitrary number of cells in a time-step. While setting up a wall to move further in a time-step comes with a corresponding decrease in accuracy, the speedup can be worth it for some cases and applications. For example, rotating cases that previously relied on the multiple reference frame (MRF) approach to keep runtimes reasonable could now be run transiently more efficiently without the significant loss of accuracy incurred by the MRF approach.
Moving Inlaid Mesh
In CONVERGE 3.0, we added the capability to include an inlaid mesh (i.e., a non-Cartesian local mesh) in your simulation domain. For certain applications, including an inlaid mesh can reduce the overall number of computational cells without compromising accuracy. However, the inlaid mesh had to be stationary. In CONVERGE 3.1, you now have the ability to implement a moving inlaid mesh. Moving inlaid meshes can be useful for a wide variety of applications, such as accurately capturing heat transfer in a piston, resolving flow around flaps and slats on an aircraft wing, or resolving flow around a moving FSI object. For applications such as these, a moving inlaid mesh could help capture the important physics of your system at a lower computational cost than a traditional Cartesian mesh.
Solid Particle Modeling
In addition to CONVERGE’s long-standing liquid-phase spray capability, CONVERGE 3.1 adds the ability to simulate solid particles. This new capability is useful for modeling a variety of phenomena and applications, such as soot, erosion, and particle entrainment and deposition.
New Release Strategy
Beginning with CONVERGE 3.2, we are implementing a new release strategy for our software. In this new strategy, clients will have access to the development branch of our code as well as several stable branches. All new features and enhancements will be implemented in the development branch, while the stable branches will receive only bug fixes. There are a number of advantages that come with this new release strategy. Clients will quickly have access to new features in our development branch, and these features will be activated through standard inputs rather than hidden inputs. At the same time, minimal changes will be made to the stable branches, which ensures the repeatability of results for clients using those versions. In addition, this strategy means that the code will be tested more thoroughly as it is being developed, since it will be available for customers to use. This also presents an opportunity for clients to provide feedback on the development branch that will enable us to shape our code to best meet our clients’ needs.
