Emobility
Engineering Excellence for Electrical Systems
As the world begins to favor more sustainable mobility options, electrical systems are becoming an integral part of the transportation industry, offering solutions including electric motors, battery packs, and fuel cells. Accompanying the rising demand for efficient, durable, and safe electrical systems is the need for a powerful CFD software that can handle the complex phenomena in these systems. CONVERGE CFD software is well suited for modeling 3D flow, heat transfer, and chemistry in complex geometries with stationary and/or moving components. Its many innovative features allow for the efficient and accurate investigation of crucial physical processes in electromobility (emobility) systems, including battery cooling, battery thermal runaway, electric motor cooling, fuel cell operation, and more.
Battery Modeling
Heat Generation & Cooling
CONVERGE includes a variety of options for modeling battery heat sources. The battery equivalent circuit (BEC) model allows you to represent the battery as an electrical network (RC circuit) and calculates the heat generation based on the electrical current using lumped model equations. A more detailed 3D electric potential solver can be employed to predict the electric potential and current field distributions and associated heat generation for direct current (DC) applications. When the electric potential solver is activated, CONVERGE solves for an electric potential solution within solid streams and porous media volumes with nonzero electrical conductivity. Through this method, CONVERGE accounts for ohmic heat dissipation, which refers to the process where electrical energy is converted into heat when an electrical current flows through a resistant material, i.e., Joule heating.
Within battery volumes, a coupled electric potential solver can be used to solve for two coupled electric potential fields, one for the anode side and one for the cathode side. The two fields are coupled through a current source term, which can be provided via a detailed electrochemical model like the Single-Particle Model (SPM). The SPM battery model can capture essential charging and discharging dynamics while reducing the computational complexity.
Effective thermal management is crucial to ensuring a battery pack’s longevity and efficiency. CONVERGE’s state-of-the-art conjugate heat transfer (CHT) modeling enables you to predict the heat transfer between fluid and solid materials to study a variety of different cooling methods, including air and liquid cooling.
Predicting and Preventing Thermal Runaway
CONVERGE can predict thermal runaway, a phenomenon that occurs when a battery cell reaches a certain elevated temperature, triggering a cascade of exothermic reactions and causing the temperature to spike. As the temperature of the cell rises, it can lose stability and rupture, releasing its stored energy. CONVERGE’s SAGE detailed chemistry solver can model the complex chain of reactions that occur during thermal runaway using Arrhenius-based chemical reaction mechanisms, like the Hatchard-Kim or Ren mechanisms. The Hatchard-Kim mechanism consists of four reactions devised for a lithium-cobalt oxide (LCO) type battery. The Ren mechanism consists of six reactions devised for a nickel, cobalt, and manganese (NCM) type battery. Chemical kinetic parameters such as activation energy, frequency factor, reaction order, and reaction enthalpy of CONVERGE’s default Hatchard-Kim or Ren mechanisms can be re-calibrated to match experimental accelerating rate calorimeter (ARC) data from a user’s specific battery case. However, there are many situations that would require generation of custom mechanisms, especially for newer battery chemistries.
In CONVERGE, you have the flexibility to use custom thermal runaway mechanisms through user-defined functions (UDFs). These thermal runaway mechanisms are created by fitting differential scanning calorimetry (DSC) data and are often battery-chemistry-specific. The DSC data fit tool in CONVERGE generates a multi-step reaction mechanism using DSC data from individual battery materials and captures the specific reaction kinetics. This solution adopts a user-specified mechanism function to fit kinetic parameters for individual reactions. The resulting mechanism will accurately reproduce intricate thermal runaway behavior within CONVERGE simulations.
When a single cell initiates thermal runaway, it can generate and transfer enough heat to cause adjacent cells to also enter thermal runaway, which can be catastrophic to the entire battery pack. CONVERGE’s Adaptive Mesh Refinement (AMR) feature dynamically refines the mesh throughout the computational domain based on the curvature of field variables, allowing the user to efficiently capture all important physical properties. By applying temperature-based AMR to the solid components of the battery, CONVERGE is able to predict a sharp and accurate thermal runaway propagation front.
Cell Venting & Ignition Risk
Thermal runaway can cause cells to vent combustible gas products into the battery pack. With CONVERGE, you can track the propagation of hot vent gas species and assess the change in solid temperatures in the area where the venting occurs. When the gas is vented, there is a chance it could ignite, causing an explosion or battery fire. CONVERGE’s detailed chemistry modeling capabilities enable you to assess battery safety by investigating the risk of ignition and subsequent combustion of vent gases. Hot solid ejecta is also present when a battery vents, which can be modeled using the Lagrangian particle approach available within CONVERGE.
Electric Motor Cooling
With demand rising for more powerful and more reliable electric motors, manufacturers are increasing the power density of their designs, necessitating better methods for removing excess heat generated by the motor. High temperatures can not only cause the demagnetization of the permanent magnets, but also break down the winding insulation and other critical components. As such, an electric motor’s power and efficiency is only as good as its cooling capacity.
CONVERGE can read electromagnetic heat source output from JSOL Corporation’s JMAG-Designer software, which allows you to obtain an accurate account of temperature distribution. CONVERGE Studio first reads in the geometry from the JMAG NASTRAN file and then assigns thermal boundary conditions or a volumetric heat generation source for additional CHT calculations.
CONVERGE has been used to evaluate various cooling methods that aim to dissipate the excess heat generated by the motor, including air cooling, water cooling, and oil cooling. In such cases, CONVERGE’s CHT capabilities can efficiently predict the temperature of the solid components to assess the effectiveness of the cooling strategy. Additionally, CONVERGE’s volume of fluid (VOF) modeling accurately captures the multi-phase flow for liquid-cooled motors. Adding AMR along with an appropriate interface capturing scheme, such as PLIC, HRIC, or FCT, can help ensure a sharp liquid-gas interface capture.
Accelerating Motor Simulations
While CONVERGE’s autonomous meshing can easily handle moving geometries, the multiple reference frame (MRF) approach further reduces computational time through a simplified modeling framework. After the user specifies a region around the moving portion of the geometry, CONVERGE treats this area as a local reference frame that moves relative to the inertial (stationary) reference frame. Within the local frame, CONVERGE solves modified continuity, momentum, and energy equations to represent the motion of the stationary geometry. This method essentially allows fluid behavior to be accurately represented without actually moving the geometry.
However, one limitation of this approach is that it cannot produce time-accurate results for certain case setups. For example, in a case with a liquid jet impinging on a rotating impeller, the standard MRF approach would show the jet entering the MRF region (the impeller) at the same location and interacting with the same blade throughout the simulation.
To bypass this restriction, you can apply rotational corrections to the MRF region. CONVERGE will recalculate the connectivity for cells at the interface between the local and inertial reference frames at each time-step. This connectivity is updated as though the cells in the MRF region are rotating, even as the actual geometry remains stationary. This creates a temporal variation of the cell face that mimics the time evolution of the moving geometry case. MRF with rotational correction can significantly accelerate liquid-cooled electric motor simulations with no notable impact on accuracy.
Fuel Cell Modeling
Polymer electrolyte membrane fuel cells (PEMFCs), also known as proton exchange membrane fuel cells, generate power by converting chemical energy released from the electrochemical reaction between hydrogen and oxygen into electrical energy. PEMFCs have high power densities with relatively low weights, and they usually operate around temperatures of 140-185°C, making them attractive for applications in the automotive industry. However, PEMFC operation can require expensive catalysts, as well as hydrogen, whose low volumetric energy density may present several modeling challenges.
CFD helps engineers improve the performance and efficiency of PEMFCs by predicting fluid velocities and pressures, species concentrations, and the relationships between fuel cell voltage and current. With its autonomous meshing capabilities, CONVERGE can effectively capture the complexity of modern fuel cell geometries, avoiding the additional pre-processing burden associated with mesh generation in many other CFD solvers. CHT modeling in CONVERGE can be used to calculate the heat transfer throughout the fuel cell stack to locate regions of low or high temperature. Additionally, CONVERGE’s multi-phase modeling can simulate the flow of liquids and gases in the reactant supply channels and gas diffusion layers, which are typically represented as porous media. This feature can help PEMFC manufacturers predict local water content and simulate liquid water transport, which are important for evaluating the performance of the fuel cell. The fully coupled solution of electrochemistry, multi-phase fluid dynamics, and heat transfer within CONVERGE allows engineers to study the activation and mass transport losses of PEMFCs, which can degrade cell performance, and design ways to mitigate them.
Speed Up Your CHT Simulations
CHT modeling is an essential component of any simulation involving heat transfer, including battery and motor cooling, combustion, and thermal management of fuel cells and power electronics. In CONVERGE, you can increase the efficiency of your simulations with available CHT acceleration methods.
Super-Cycling
In many CHT simulations, temperature evolution in the solid region occurs more slowly than in the fluid domain. The difference in time-scales is often problematic for CFD engineers since extremely long simulations are computationally expensive. CONVERGE overcomes this problem with super-cycling, which freezes the fluid solver while the solid solver progresses toward steady state. This method allows you to quickly arrive at a steady-state temperature solution within the solid domain.
Fixed Flow
In simulations where the solution in one stream is considered quasi-steady, CONVERGE offers the fixed flow method, which can be useful for transient cases with large simulation durations. The quasi-steady stream is frozen in a cyclic manner, allowing the solutions in solid streams to progress quickly. This scheduled progression efficiently simulates cases that would otherwise be prohibitively expensive.
Cross-Stream Synchronization
The cross-stream synchronization feature allows users to apply different time-steps to different streams, essentially accelerating transient simulations where simulation time-scales may be highly variable across separate streams. For example, in CHT simulations of battery thermal runaway, fluid streams require much smaller time-steps than solid streams. With cross-stream synchronization, users can specify a time interval to synchronize both solid and fluid streams. This approach significantly saves computational resources that would otherwise be spent on solving the slow solid stream. Cross-stream synchronization can also be applied to steady/transient coupled simulations, offering a more implicit way to couple steady and transient simulations. A schematic depicting a broad overview of how this feature works is shown in the figure below.
Autonomous Meshing
CONVERGE’s autonomous meshing is designed to simplify your workflow and reduce total time-to-solution. At runtime, CONVERGE automatically generates a Cartesian mesh based on a few simple user-defined parameters, effectively eliminating all user meshing time. With a novel cut-cell technique that guarantees an accurate representation of your geometry, no matter how complex, this helpful tool will redefine your CFD simulations.
For moving geometries, CONVERGE regenerates a stationary mesh at each time-step to accommodate the motion without moving or deforming the mesh. This approach avoids numerical diffusion, which may occur with moving meshes and, overall, results in a more accurate solution.
What’s more, Adaptive Mesh Refinement (AMR) intelligently adds and removes cells throughout the simulation domain to resolve complex phenomena such as the flame front in a battery pack ignition simulation or a sharp liquid-gas interface for multi-phase flow simulations. You can also adjust your AMR settings to specifically assess gradients in parameters such as velocity or temperature.
Many emobility applications, such as electric motors, have complex and/or moving geometries, making CONVERGE and its unique meshing abilities well suited for capturing the important physics both accurately and efficiently.
Adding Grid Refinement
In certain simulations, such as an electric motor with small rotor-stator gaps or a thermal boundary layer around electrochemical cells in a battery pack, adding finer local meshes to a coarse base grid can make the simulation more efficient. CONVERGE offers two ways to incorporate a locally refined region within a coarser computational mesh.
CONVERGE’s fixed embedding feature allows you to refine the grid at specific locations in the domain where a finer resolution is critical to the accuracy of the solution, while keeping the rest of the grid coarse. Another option is inlaid meshing, which allows you to import or construct a local non-Cartesian mesh with prescribed surface information for flow-through boundaries. This technique consists of a finer user-generated mesh “inlaid” into the autonomously generated background mesh, where connectivity between the two meshes remains physically consistent at the interface.
Both fixed embedding and inlaid meshing can be activated and deactivated during the simulation, but they are non-adaptive strategies, meaning the mesh resolution is fixed in place and will not change while active. While fixed embedding can be more straightforward and user-friendly to set up, a user-defined inlaid mesh can achieve a more optimal cell count using directional refinement.
