Search results

Aging affects most things on Earth, and batteries are no exception to this phenomenon. In a very peculiar way, batteries behave like "living" beings; their particles flow between two electrodes, there are chemical reactions, and even mechanical changes, such as an effect similar to "breathing" (expansion and contraction of the electrodes due to the intercalation and deintercalation of lithium during charge/discharge cycles). They are simply in operation, which generates natural wear and tear. You Can't Stop Aging: Electric Vehicle Fleets Will Age (and so will their batteries) However, when people consider buying an electric car, the number one criterion for them is range, followed by price and charging logistics (infrastructure and charging time). Concern about battery life only appears in 6th position. Today, this ranking may not be surprising, considering that most electric vehicles are bought new and aging seems to be a rather technical topic, with a wide variety of possible evolutions in battery life depending on the use of the electric vehicle. But, as electric vehicle fleets age (and with them their batteries), the used car markets begin to grow and, suddenly, for good resale prices, the lifespan and health of the batteries will certainly increase in importance (as predicted). Similarly, recycling battery cells that have reached the end of their useful life is still a very expensive and energy-intensive activity. And therefore, longer-lasting and more sustainable battery cells are already a competitive factor for those who design and sell electric vehicles and batteries. So, the time has come for OEMs and battery manufacturers to understand battery aging and develop cell designs that provide maximum lifespan. Engineers must understand not only when, but also where aging mechanisms occur. Using simulation with aging models can significantly help accelerate the prediction of the degradation trend of a given battery. Typically, 1D level simulations are used in this case, as they allow for very fast execution and can produce years of simulated data in a few hours. Throughout this article, it will be exemplified how this can be done in a simulation environment where physical phenomena are modeled, but first let's learn a little more about the causes of battery aging. The Toxic Ingredients That Cause Battery Cell Aging So, what triggers and affects aging? Battery aging has its root causes in several factors. First, unsurprisingly, time: whether the cell is being used or remains idle, time is at work allowing some internal chemical reaction to induce some performance degradation. Second is temperature: temperature has a significant impact on the battery lifespan degradation process. Storage and use at high temperature (high range of safe temperature limits) would accelerate aging. Low temperatures are better, but combined with fast charging can be recipes for other degradation effects. This leads to the third main criterion, the current applied to the cell. Basically, referring to the type of load applied to the battery. If it is used gently with smooth and low power demands, the current applied to the cell will be smooth and slowly affect aging. However, if the battery is used more aggressively, with more frequent fast charging, particularly under low temperature conditions, the accelerated degradation mode will be activated. A deeper analysis of battery cell degradation mechanisms What happens within the battery due to these effects is a combination of several degradation mechanisms: Solid Electrolyte Interface film growth: This is the slow growth of a thin, porous layer on the surface of the active material, which consumes Lithium atoms to grow. As it grows the inventory of available lithium, used for the cell operation, decreases, reducing the cell’s capacity. Also, the thickness of the SEI film creates a barrier to the Lithium ions and electrons trying to go in and out of the active material, which increases the cells’ overall electrical resistance Lithium plating. In this case, there is formation of lithium metal film on the surface of the active material, which also consumes the lithium inventory impacting the cell capacity. Loss of active material by dissolution: Active material responsible for storing lithium is dissolved into the electrolyte due to some undesired side reaction. The loss of this active materials further decreases the cell capacity. Loss of active material by mechanical cracking . The Lithium intercalation and de-intercalation process generates at each cycle some mechanical stress. Overtime parts of the active material can break down and be separated from the main electrode. This has the effect of losing ability to store lithium and further decreases the cell capacity. The bottom-line consequences of these effects are simple, your battery’s capacity will decrease, reducing your vehicle range compared to its brand-new range. And it will be less able to sustain aggressive power demands, reaching more rapidly the lower and maximum voltage safety limits, leading to the battery’s shut down. Aging takes time – that engineers don’t have That's why battery and vehicle manufacturers dedicate time and effort to characterizing these aging phenomena. But here's a challenge: the effects of aging can only be observed after several years of operation. Therefore, as you can understand, conducting tests to capture the correct degradation behavior requires an enormous amount of time and money to test the battery over the years of operation! Of course, there are some accelerated aging testing techniques, but the first results can only be seen after at least 6 months of accelerated aging tests. But to gain a competitive advantage, engineers analyzing these aging challenges need more detail; they need to further optimize the battery cells and understand not only when, but also where the aging mechanisms occur, so that they can better address degradation problems locally. EV battery cell formation (the initial charge) is a critical manufacturing step with respect to battery cell aging risks (Image: Chroma ATE). Inspection & Identification The first stage of the process. Battery cells are inspected and identified before entering the formation cycle. Formation This is where the initial charging of the cell happens — known as the “formation” process. A critical step that defines the cell’s electrochemical properties and directly impacts its lifespan . Ambient Aging Cells rest in ambient temperature conditions. This step allows the materials inside the cell to stabilize after formation. High Temperature Aging Cells are kept at elevated temperatures to accelerate aging and detect early defects. Ensures only stable cells move forward in the production line. OCV & ACR Testing Electrical testing to measure: OCV (Open Circuit Voltage)  – voltage when the cell is not under load. ACR (Alternating Current Resistance)  – internal resistance of the cell. These tests assess the performance and quality of each cell. Sorting Cells are classified based on results from electrical and aging tests. Cells with similar performance are grouped together to form consistent battery modules or packs. And it's not only aging that needs to be studied during operation: equally relevant is the initial charging process, known as formation, which is the critical final stage of manufacturing before the cells are shipped. It forms the crucial protective layer of the Solid Electrolyte Interface and therefore has a huge impact on the subsequent lifespan of the battery. Battery aging simulation There are several approaches to leveraging simulation to predict aging and the formation process. Firstly, our Simcenter Amesim systems solution, using 1D models, can be extremely efficient in rapidly generating years of aging simulation data under various operating conditions. The main advantage here is time acceleration. Physics-based aging models in Simcenter Amesim have been available since version 2410, in addition to the existing empirical aging models. In this type of simulation, each cell is represented by blocks that describe its electrical and thermal behavior—capacity, internal resistance, and heat exchange with the environment. By connecting multiple cells in series and parallel, it is possible to predict how performance and temperature evolve over time, simulating battery aging and allowing for design adjustments before moving on to more detailed 3D analyses in Simcenter STAR-CCM+ . Second, to address the need for spatial information, Simcenter STAR-CCM+ 's 3D Cell Design solution can predict aging evolution in a 3D cell geometry with resolved electrode layers. Of course, in this case, the execution time is much longer than in 1D simulations, but the user will have access to local information about where aging occurs and can mitigate these effects by changing the design or operating conditions. Thirdly, it is possible to combine 1D and 3D simulations. The 1D simulation is used to generate the very long-term aging simulation over years of physical time. Users can then extract from this discrete point the cell's State of Health (SOH) over the aging period, for example, every year. This SOH for each year can then be a starting point for a 3D simulation, where the cell is aged only for a short period, for example, 1 month of physical time, but long enough to generate the distribution of the various aging mechanisms, such as Solid Electrolyte Interphase (SEI) growth or lithium plating, as implemented in more recent versions of Simcenter STAR-CCM+ . Obviously, the 1D and 3D aging models are coupled with thermal models to capture the thermal effect on the evolution of degradation mechanisms. Finally, 3D simulations can be used to assist in predicting the initial Solid Electrolyte Interphase (SEI) layer during the manufacturing formation process. In fact, the SEI growth model can be used in the first charge of a battery cell and predict the growth of this critical protective layer. The 3D Cell Design feature can then help the user evaluate the uniform evolution of the SEI layer growth and determine the optimal point at which the layer is sufficiently thick and the amount of lithium consumed to generate it. This will help further refine the estimate of cyclable capacity. High fidelity battery aging simulation with Simcenter STAR-CCM+ Aging through parasitic side reactions with Sub-grid Particle Surface Film model Available since the release of Simcenter STAR-CCM+ 2406 , the “Sub-grid Particle Surface Film” model in the Battery Cell Designer allows simulating the cell's response to a duty cycle in relation to two of the main degradation mechanisms: The growth of the Solid Electrolyte Interphase (SEI) film The growth of the lithium metal plating film An Active Material Particle, presented at NordBatt Conference These are both parasitic side reactions which occur during the cell operation. Lithium plating is the deposition of Li-metal on the particle surface. And SEI is the film created from the reaction between the particle and the electrolyte. Due to the side reactions, the amount of cyclable lithium reduces, you can simply track the remaining lithium in the electrolyte and the active material. This should allow to check the effect on the capacity. The film resistance area (resistivity times thickness) is also a field function which can be tracked and contributes to the overall internal cell resistance. Mechanical-induced degradation with the Sub-Grid Particle Aging model Simcenter STAR-CCM+ includes "Sub-Grid Particle Aging," which focuses on degradation effects of a mechanical nature. In this case, the loss of active material due to mechanical stresses is characterized by alternating stresses during charging and discharging, i.e., the cyclic insertion and extraction of lithium from the active material particles, which can lead to the formation of cracks in the electrodes. This can cause loss of electrical contact and reduction of usable active material, leading to an overall loss of cell capacity and an increase in internal resistance. Active material particles undergoing surface cracking and loss of active material There are two types of cracks forming, represented with two model options under the “Sub-grid Particle Aging” model: First one is the “Loss of Active Material” model. It is characterized by the cracking of particles or electrode “blocks”, leading to an electrical contact loss of active material particles, making those particles electrochemically inert and no longer participating in the electrochemical reactions. These particles represent therefore a loss in cell’s capacity The second effect is the “Surface Crack Growth” model. The cyclic insertion and extraction generate cracks within the particles themselves. Those cracks expose a new surface for the Solid Electrolyte Interface (SEI) to grow, leading to Lithium consumption and therefore an overall capacity loss and internal resistance increase. Note that this model option is compatible with the “Sub-grid Particle Surface Film” model enabling the SEI growth effects simulation. Also note that, some publications on the topic suggest, that the tortuosity should increase when the surface cracks grow. A trustworthy battery aging simulation framework The abovementioned aging models were validated against experimental measurements generated during the EU commission funded project MODALIS² , which was focusing on developing physics-based aging models for the latest generation of Li-ion battery cells. This work was performed with key industrial partners specialists in the field of batteries, such as a cell maker, cathode supplier and electrolyte supplier. All that said, thanks to high physical modeling fidelity and the unique three-dimensional implementation of the models, these aging models offer you the ability to localize areas of the cell which most impacted by all types of aging. This is in theory. So let’s look at those models in action. Simulating aging cycles in 3D This first example was presented at the NordBatt conference in 2022 by my colleague Stefan Herberich from SIEMENS . A prototype cell used in the EU-funded MODALIS² project was used, and the cell is tested over several cycles with aggressive aging conditions to locate the weak areas where degradation is most dominant. The cell considered consists of 15 electrochemical layers. The discretized cell is shown below, along with some results. In total, there are approximately 200,000 finite volume cells. In particular, the thickness direction is discretized using 10 cells per anode and cathode layer and 2 cells for the separator and current collectors. The drive cycle consists of the following steps: first, charging is performed with constant current (CC), applied at a rate of 2C. C-rates indicate the ratio between the charging current and the battery capacity—at 1C, a fully discharged battery (0% state of charge, or SOC) is fully charged in 1 hour; at 2C, the current is doubled, and charging is completed in approximately 30 minutes. If the voltage exceeds 4.2 V, the process switches to constant voltage charging mode, remaining at 4.2 V until the state of charge reaches 95%. The 4.2 V limit is reached quickly. Then, the battery remains at rest for a little over 3 minutes and is then discharged to 60% state of charge, also at a rate of 2C. After another rest period, the complete cycle is repeated ten times. Interpretation of results The study provides insights into the effects of the two aging mechanisms that occur: SEI growth and the influence of lithium plating side reactions. The images show the average thickness of the SEI layer around the particle and the equivalent average thickness of the plated lithium on a particle, respectively. The corresponding results were observed on the anode plane and in a cross-section in the direction of the cell thickness. In addition to the analysis of the SEI, this study also provides important information about LAM (Loss of Active Material), which refers to the degradation or inactivation of the electrode material that participates in the electrochemical reactions. In the plane: the thermal boundary conditions are such that the highest temperatures are observed in the center of the battery cell. At this location, the temperature dependence of multiple material parameters leads to higher SEI growth rates. LAM is pronounced near the battery tabs, where the highest rates of voltage variation are observed. In thickness: As expected, SEI and LAM growth are greater near the separator. The operating conditions are such that the lithium metal, with an initially specified homogeneous profile, is dissolved more quickly than deposited, especially near the separator. SEI during the formation step The second study will be on SEI during the initial charge, also known as formation. Using the results presented in “Andrew Weng et al. 2023 J. Electrochem. Soc. 170 090523”, Simcenter STAR-CCM+ and the “Sub-grid Surface Film” model were used to replicate this study. The article describes the formation of SEI, i.e., the accumulation of a passivation layer on the graphite anode of a battery during the first charge cycles. The film layer is formed due to a side reaction of the solvent components S, ethylene carbonate (EC) and vinyl carbonate (VC), with Li+, which produces the film components P, lithium ethylene dicarbonate (LEDC) and lithium vinyl dicarbonate, and gaseous byproducts Q. Only the first 4 hours of the formation process were simulated, which is when the rapid dynamics occur and the transition from the kinetically limited regime to the diffusion-limited reaction regime takes place. The results reasonably correspond to the reference: The results demonstrate the ability to use Simcenter STAR-CCM+ in an approach to understand the SEI formation process, but also to be able to better control it and brings the potential to reduce its overall duration, which in some cases can last up to ~20 days. Want to understand how to predict and mitigate battery aging with high accuracy and efficiency? Schedule a meeting with CAEXPERTS and discover how Simcenter Amesim and Simcenter STAR-CCM+ solutions can revolutionize the development of longer-lasting and more sustainable cells for electric vehicles. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Build a scalable Battery Energy Storage System (BESS) and achieve high ROI This post focuses on the crucial role of energy storage in promoting corporate sustainability and profitability. By integrating BESS with renewable energy sources, companies can achieve significant cost savings, reduce their carbon footprint, and boost long-term profitability. We will explore how BESS industry leaders are creating digital plants, increasing flexibility, and building a competitive advantage in a rapidly changing market. If your company is ready to lead the transition to BESS, this is your roadmap. System Simulation Plays a Crucial Role System simulation plays a crucial role in the techno-economic evaluation of battery energy storage systems (BESS) in the energy sector, especially when integrated with renewable energy sources such as wind turbines and solar photovoltaic (PV) systems. Various use cases covered by BESS Here are some key aspects: Balancing Power Generation and Consumption : Peak Shaving and Load Shifting : By simulating different load profiles, BESS can be optimized for peak shaving (reducing peak demand) and load shifting (moving energy consumption to off-peak times), which can reduce energy costs and improve grid efficiency Grid Stability : Simulations can assess how BESS can be used to balance intermittent renewable energy generation with grid demand, enhancing grid stability and reliability Integration with Renewables : Energy Management : Advanced energy management strategies can be simulated to coordinate the operation of BESS with renewable generation, ensuring that energy is stored and dispatched in the most efficient way. Including the weather conditions. Energy Price Evolution : Forecasting and Optimization : System simulations can model future energy price scenarios, helping to optimize the operation of BESS for energy arbitrage (buying low, selling high). This ensures that the BESS is used in the most cost-effective manner OPEX/CAPEX : Cost Analysis : Simulations can provide detailed cost-benefit analyses, including capital expenditures (CAPEX) and operational expenditures (OPEX). This helps in understanding the financial viability and payback period of BESS projects Degradation Modeling : By simulating the degradation of battery cells over time, it is possible to estimate maintenance costs and replacement schedules, which are critical for long-term financial planning Overall, system simulation provides a comprehensive framework for assessing the technical and economic feasibility of BESS projects, helping stakeholders make informed investment and operational decisions. We will explore the initial phases of a Battery Energy Storage System (BESS) project together, focusing on some technical and economic assessments for success (OPEX/CAPEX, energy price trends, load balancing, return on investment), and moving through the different stages with Simcenter System Simulation : To calculate your customer's electricity bill Considering some weather forecasts From renewable energy (solar PV) Battery Electric Storage System (BESS) optimization and control strategy To typical results in operations based on realistic scenarios The use case here is a food processing plant near Lyon, France. Some effort was devoted to modeling the solar photovoltaic (PV) system integrated with the BESS and surrounding consumers. With its load and heating system represented over a one-year period (January to December), the digital twin considers solar PV BESS operations with different electricity tariff structures and PV or BESS unit costs. Therefore, it addresses the technical and economic value of adopting BESS in dynamic tariff structures. Although the case study was conducted in France, it is important to highlight that even more impressive results could be obtained in countries with higher solar incidence, such as Brazil. High irradiance throughout the year significantly increases the potential for photovoltaic generation, increasing BESS efficiency and reducing the payback period. Thus, the combination of high solar resources and the application of advanced simulation strategies makes the Brazilian scenario especially promising for energy storage and renewable energy integration projects. Better Design for Operational Excellence As the BESS economy gains unprecedented momentum, companies are racing to meet the growing demand for clean energy. However, scaling production while remaining profitable, sustainable, and resilient poses a formidable challenge. BESS producers and equipment manufacturers must overcome fragmented data systems, high energy costs, and supply chain complexities to stay ahead. Companies striving for operational excellence are already turning these challenges into opportunities. By leveraging digital twins and advanced simulations, they are optimizing processes, reducing costs, and improving scalability. The energy industry needs BESS to save money and reduce carbon emissions. Digital Twin for an Industrial Facility This is a distributed BESS digital twin to predict and optimize system performance with multiphysics. It includes consumers (food processing facility: 20°C) with the heating system and customer load, grid connection, solar PV with solar panels, stationary batteries, as well as a smart controller based on weather conditions and energy price fluctuations. BESS Digital Twin in Simcenter Amesim In the image above, the Heating System, Consumer Load, and Grid Energy blocks were modeled in Simcenter Amesim based on tables containing actual operating data from the industrial facility and predefined formulas for calculating grid energy consumption and costs. These blocks represent the system's energy behavior under different load conditions, providing a reliable basis for analyzing the performance and operational costs of BESS systems. The Solar Panel and Battery blocks use simplified physical models to represent the actual operation of these devices. Photovoltaic generation is simulated using historical solar irradiance and ambient temperature data, while the battery model was parameterized based on information from technical catalogs, allowing for the prediction of load states, efficiency, and available capacity. These models capture the system's energy dynamics. Finally, the Intelligent Control block integrates all this information to enable real-time decision-making. Thus, the control optimizes the energy flow between generation, storage, and consumption, seeking to reduce costs and improve overall system performance. By running the simulation, users can access all variables from the different subsystems. Thus, complete information is available, from consumer load [kW] to the electricity bill over time [€] (€1 to $1). The evolution of electricity prices throughout the year, with their daily fluctuations, was considered. Below, two price trends are shown on January 1 and July 19, to obtain information on minimum/maximum prices at different times of the year. Typical results obtained with the BESS digital twin The solar panel includes your GPS location, turbidity factor (the effect of particles, similar to smoke in the air), or cloud cover factor (for weather conditions). Cloud cover factor changes depending on weather conditions At the same time, the outside temperature changes are included from a known database, allowing users to assess its impact on the heating system, considering the factory indoor temperature setting. Temperature Evolution (ceiling, outdoor, indoor) and Air Conditioning Power This analysis allows you to calculate associated information, such as air conditioning power [W] or the energy consumption of all surrounding subsystems. This allows users to obtain a realistic power evolution to evaluate the balancing mechanism and optimal control strategies to implement in their BESS system. BESS Macroanalysis with Realistic Scenarios The industrial unit model allows for mass exploration. Analyses are completed in just a few minutes, opening the door to long and complex scenarios. Energy generation and consumption [kW] for all subsystems (the battery is inactive here) Users can practically evaluate the energy generation, storage, and consumption of all subsystems. Meanwhile, the intelligent control system manages the Energy Management System (EMS) to distribute the energy, store it in the BESS, or deliver it to the grid. Since all changes are intermittent or dynamic, a system simulation tool, such as Simcenter Amesim , is necessary to optimize sizing and control strategies. Finally, the user can access variations in energy flows over time. For example, you can check the energy generated by solar panels or brought in by the grid, as well as the energy supplied by the battery. This corresponds to the energy needed by the load, while some small levels of energy are taken from the grid or returned to the battery during off-peak periods when demand is low. Energy Flow Variations Over Time This is a major achievement! We can already observe good results thanks to the digital twin with Simcenter System Simulation . But it's possible to go much further, more technically and economically. See how you can save US$1 million over 20 years while simultaneously reducing a huge amount of CO₂ emissions, down to -17 tons of CO₂ equivalent. Save US$1 million and tons of CO₂ equivalent We will now address the business and decarbonization aspects, with the goal of demonstrating how it is possible to create a scalable forecast for BESS systems, in order to measure and replicate significant successes. The digital twin of the food processing unit is equipped with metadata to produce the relevant economic KPIs (key performance indicators) to ensure its monetization, return on investment (ROI), or payback through CAPEX (capital expenditures) or OPEX (operating expenses). Operating Costs [$k] during the Scenario The reference is the electricity bill without solar panels or BESS. It amounts to US$103,000 paid over one year. By installing the solar panels and BESS, the new electricity bill, which is now US$33,000 after one year, can be captured, with an investment of US$625,000 for the photovoltaic system and US$77,000 for the BESS. This corresponds to a savings of US$70,000 per year in OPEX, thanks to the installation. Discounting CAPEX costs, a benefit of US$698,000 is obtained after 20 years of operation. CAPEX costs are reimbursed after a 10-year payback period. Knowing that the value doubles every 15 years due to the interest rate, it can be assumed that the actual savings will reach US$1 million after 20 years. Please note that this is a preliminary calculation that shows the potential, while things like inflation and maintenance costs are not covered, which is good for a first estimate. Profitability [$k] including payback [year] Now is the right time to optimize sizing and extract maximum value from the new installation. Discover the maximum benefits and best returns in just a few clicks. A batch study was configured to vary selected parameters, defined as the number of solar panels (366, 488, 610) and the number of battery racks (0, 100, 150). It was observed that the payback period can be reduced to approximately 9 years (-11%) in the most favorable configurations, while other options can extend it to up to 12 years (+20%). Comparison of return on investment [year] depending on the number of solar panels and the number of racks Finally, for the Earth's good health in relation to climate change, it is also essential to consider the reduction of carbon emissions thanks to renewable sources, the BESS system, and smart control strategies. CO₂ emissions from the grid, load, and heating, as well as the total CO₂ reduction per year A significant reduction of 17 tons of CO₂ equivalent per year was achieved. This result represents a significant contribution to the sustainability process through decarbonization. All these achievements were achieved using the digital twin through Simcenter System Simulation . Going Further You can even go a step further, introducing a new paradigm with grid supervisory control. The latest and most innovative technologies allow the combination of artificial intelligence (AI), weather forecasting, and streamed data. This offline digital twin is converted into an executable digital twin that connects real-time performance data with accurate and well-orchestrated plant information and simulation tools, allowing you to troubleshoot critical system situations (surges during switching, etc.) or benefit even more from price and CO₂ reductions. What a great prospect! Grid Supervisory Control, Combining AI, Weather Forecasting, and Streaming Data In short, owner-operators in the global BESS business have a historic opportunity to expand their business and market share in the coming decades. The companies that will emerge as leaders in the delivery sector will be those that can overcome the complexity of BESS and turn it into a competitive advantage. System simulation definitely helps you succeed in your BESS journey thanks to digitalization, system integration, and intelligent controls. Schedule a meeting with CAEXPERTS and discover how to transform the potential of your BESS project into real results. Our experts will show you how the use of digital twins and system simulation can optimize sizing, reduce OPEX, and accelerate return on investment (ROI)—all while your company advances in the energy transition and decarbonization. Take the next step toward efficiency and sustainability: contact us today. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Challenge: achieve fast and accurate virtual prototyping The optimal multi-attribute design of transmissions remains a critical challenge for drivetrain engineers, gaining even greater significance in the era of electric vehicles. Today’s automotive industry demands powertrains that excel in multiple aspects: minimizing noise and vibrations, maximizing power density and efficiency, while ensuring unprecedented durability performance. For this purpose, transmission engineers need to design innovative transmissions that meet the multi-attribute performance criteria. Computer-Aided Engineering (CAE) allows engineers in industry to virtually prototype and optimize their next product optimization. Simulations are essential to accurately predict and optimize component behavior and the system-level transmission performance throughout the development cycle. Powerful physics-based models and simulation capabilities are available, which accurately model real-life products, perform predictive simulations and optimize the product’s performance for statics, dynamics, aerodynamics, acoustics, durability, etc. These physics-based models, for example Finite Element (FE) models, are successfully adopted in industrial development workflows. For high-fidelity simulations (e.g. detailed FE contact simulations), the model fidelity must increase to allow accurate predictions, which unfortunately also increases the model’s computation time and cost. In CAE, Artificial Intelligence (AI) and Machine Learning (ML) technologies are revolutionizing physics simulations. Among the advancements, surrogate modeling for physics simulations experiences rapid growth thanks to advanced AI techniques and architectures. Surrogate models, or reduced-order models (ROMs), provide efficient alternatives to computationally expensive high-fidelity simulations (such as detailed FE contact simulations). These AI/ML-based surrogate models, capable of providing fast and accurate predictions while maintaining high fidelity, unlock the potential for extensive and automated design space exploration, enabling engineers to efficiently evaluate thousands of design variants in a fraction of the time required by traditional simulation methods. Gears, as critical components in transmission systems, particularly benefit from such advanced simulation approaches, as their design requires careful consideration of deformation, contact and bending stresses, and durability. This blog post introduces a novel gear design analysis method3 that combines the strengths of FE and AI/ML models to achieve efficient and accurate gear stress prediction, demonstrating how this technology can be practically implemented in industrial workflows. Solution: combine powerful high-fidelity models with AI/ML The innovative gear stress analysis method3 exploits the powerful predictive simulation capabilities of FE models to generate accurate simulation data, which is then used to train an AI/ML surrogate model. Once trained, the efficient AI/ML surrogate model can accurately predict gear stresses for unseen combinations of design parameters. This combines the strengths of FE and AI/ML into a novel AI/ML-accelerated gear stress analysis capability that provides accurate predictions quickly. The simulation workflow for comprehensive gear analysis integrates the capabilities of Simcenter 3D and Simcenter Nastran software solutions, while Simcenter HEEDS serves the dual purpose of automating data generation for AI/ML model training and orchestrating multi-objective optimization studies that make use of the AI/ML model calculations. Figure 1 illustrates the automated data generation workflow, in which Simcenter HEEDS orchestrates the physics-based simulation process. The workflow streamlines the creation of gear datasets, while incorporating elements from Simcenter 3D Motion Gear Design Optimization methodologies. For each design, first the macrogeometry and microgeometry of the gear pair are evaluated before the nonlinear FE analysis (NLFEA) module generates the FE model and uses Simcenter Nastran for automated FE-based gear contact. Figure 1: Automated gear data generation workflow, involving Simcenter HEEDS as orchestrator. This workflow adopts high-fidelity physics-based simulation models, bringing several advantages to design engineers: highly accurate predictions of real-world behavior through detailed modeling of physical phenomena (such as the nonlinear contact mechanics in gear mesh), and a deep understanding of complex interactions and phenomena that could be missed in simpler models. These detailed models provide a reliable source of training data for AI models and AI model validation. To accelerate gear stress and durability analysis, our approach replaces the traditional NLFEA-based contact analysis by an AI/ML surrogate model, trained on NLFEA contact analysis results, to predict the altering gear blank and tooth root stress throughout the gear meshing cycle. The resulting AI/ML-based ROM offers significantly reduced computation times compared to the full order NLFEA-based simulations, enabling rapid design iterations. Embedding the AI/ML model within the standard gear design simulation workflows creates an AI/ML-accelerated process that remains highly accurate, while opening doors to multi-objective optimization studies (typically unachievable via traditional, resource intensive, FE models). Two additional solution elements are introduced to realize the AI/ML accelerated workflows: Simcenter Reduced Order Modeling8 that enables the creation and deployment of ROMs from simulation and test data. An upcoming Simcenter AI technology for 3D surrogate modelling that leverages operator learning techniques. Results: realize fast and accurate gear stress analysis A solution workflow has been introduced, aimed at combining the power of high-fidelity models with AI/ML models to achieve AI/ML-accelerated gear stress analysis. This section introduces a gear use case to verify the workflow vs. the objective to achieve efficient yet accurate gear stress analysis. The gear design use case is described in detail below and has four design parameters: the normal pressure angle, in range [18°-22°], the addendum coefficient of the Pinion, in range [1.00-1.30], the addendum coefficient of the Wheel, in range [1.00-1.30], and the profile shift coefficient of the Pinion, in range [0.35-0.70]. Using Simcenter HEEDS and Simcenter Nastran , 81 gear designs were created, varying the four design parameters through a level 3 full factorial DOE. The data set is split randomly into 64 designs as training set and 17 designs as test set. A transformer-based operator learning AI/ML model is trained to predict gear surface stresses, given the surface geometry and contact forces as model inputs. In this first study, we chose to have the AI/ML model learn the full 3D signed von Mises stress field, which captures both compressive and tensile behavior. The signed von Mises stress is created via postprocessing of the full 3D tensor at each node, which is stored in the NLFEA-based contact results. The model focuses on surface nodes where failures typically initiate (tooth root and contact regions), optimizing computational efficiency while maintaining accuracy for critical stress predictions3. Three approaches are evaluated for the gear design use case: FE Reference: NLFEA-based contact simulations (Reference), ML with FE Forces: AI/ML predictions using NLFEA contact forces, ML with Motion Forces: AI/ML prediction using Simcenter 3D Motion contact forces. Multibody-based gear contact forces (computed via Simcenter 3D Motion ) are used in this study as they provide a computationally efficient alternative to expensive nonlinear finite element analyses, while still maintaining sufficient accuracy in predicting contact force distributions – an aspect that will be crucial for future design optimization workflows. Figure 2 presents an analysis of the full stress field using the three approaches. The stress field predicted by the AI/ML surrogate model with FE-based contact forces as input (Fig. 2b), shows to be near-identical to that of the reference FE-based results (Fig. 2a), while the stress field predicted by the AI/ML surrogate model with Motion-based contact forces as input (Fig. 2c) shows only minor deviations from the reference results. These findings are confirmed by Figure 3, which presents the results for the tooth root stress during five mesh cycles for a point located along the middle of the face width and in the middle of the tooth root arc. Figure 2: Von Mises (signed) stress fields, computed from a) NLFEA-based contact simulations, b) ML model with FE-based forces (input), c) ML model with Motion-based forces (input), for a pinion gear example case. Figure 3: Comparison of tooth root stress for a point (middle flank, middle root arc) over the course of 5 mesh cycles (created based on the stress of 5 teeth during 1 mesh cycle) for a pinion gear example case Figure 3. Though data generation and AI/ML model training require an initial time investment, these are one-time offline processes. The resulting trained model enables rapid predictions, ideal for efficient design space exploration and optimization. The AI/ML model that was used in this study requires about 0.1 seconds to compute the full stress field for one point in the mesh cycle, while one NLFEA contact simulation requires on average about 5 minutes. Knowing that a full mesh cycle requires about 20 to 30 angular configurations per gear pair and that an industrial transmission has multiple gear pairs for which hundreds of variants are explored, the inclusion of the AI/ML surrogate models can truly accelerate gear design optimization. Conclusion and outlook This study demonstrates the successful integration of AI/ML surrogate models into gear design workflows, achieving both speed and accuracy in stress prediction. The AI/ML approach delivers results that closely match traditional nonlinear finite element analyses while being multiple orders of magnitude faster. This dramatic speed improvement, combined with maintained accuracy, opens new possibilities for comprehensive design space exploration and optimization of transmission systems. The developed workflow, supported by Simcenter tools , and validated through practical case studies, effectively demonstrates the industrial viability of AI/ML-accelerated gear and transmission design. References D. Park, A. Rezayat and Y. Gwen, “Gear design optimization for multi-mesh and multi-power flow transmissions under a broad torque range incorporated with multibody simulations”, in VDI International Conference on Gears 2022, Munich, Germany, 2022. M. Vivet, J. Melvin, S. Donders, “Advancing bevel gear contact simulation towards quiet transmissions”, Simcenter Blog, August 19, 2024. M. Vivet, D. Park, A. Scheuer, “AI/ML-accelerated gear durability analysis within gear design optimization”, in VDI International Conference on Gears 2025, Garching near Munich, Germany, September 10-12, 2025. Schedule a meeting with CAEXPERTS and learn how to accelerate gear stress analysis with solutions that combine high-fidelity simulation and Artificial Intelligence. By combining high-fidelity models and AI/ML, we help your company reduce development time, increase analysis accuracy, and accelerate innovation in transmission systems. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

The industry has an actual deadline to meet: 2050. Since the adoption of the International Maritime Organization (IMO) Strategy on Reduction of Greenhouse Gas (GHG) Emissions from Ships in 2023, the maritime industry has been looking at ways to reach its goals on time. Whether modifying existing fleets or building new ships, the engineering challenges arising to comply with environmental guidelines are huge. Naval architects and design engineers need to make radical decisions faster. To do so, they turn to simulation. In this maritime context, simulation is no longer a nice-to-have tool, it’s essential for exploring every aspect of vessel performance and numerous design options. To meet your targets, you need CFD simulation tools that make it easy to set up complex cases, embed best practices, and enable automation. You also need tools that are fast to run and contain the advanced multiphysics features needed for marine simulation. The good news is that you can do all of that and more with Simcenter! Should we introduce it? Simcenter STAR-CCM+ has a full suite of tools to completely automate processes and workflows and create simulation templates. These templates allow for repeatable workflows that can be created by experienced analysts and passed on to design engineers to make engineering decisions. Simulation templates can eliminate most of the manual processes involved in building a CFD simulation via a parameterized template model by boiling down the simulation to its minimal set of inputs. From there, all meshing calculations, boundary conditions, solver settings and post-processing can be applied such that simulations are repeatable and embedded with your organization’s best practices. The templates utilize tools all built into the software, so they can be updated when desired; they are not black box tools you get from your software provider! One such template is the virtual tow tank (VTT) template. Let’s explore it a bit more, shall we? Setup time: it went so fast, I didn’t see it happening… Setting up a marine CFD simulation requires a lot of steps: importing CAD, defining mass and hydrostatic properties, creating a suitable mesh, defining solver physics… Not here. The parameterized template model automates all these settings and calculations. There is no need for scripting and the file can be reused for multiple simulations and shared with others. So, using a template ensures consistency and best practices across all simulations and the repeatability of results. Marine-focused templates guide you through set up and running and ensure your best practices are applied to every simulation. Okay… What’s next? Meshing? Traditional CFD approaches require several hours of manual setup. What size should the mesh be? How do I suitably build my boundary layer mesh? And where should I place the appropriate refinements? If you think about it, all these decisions are based on the vessel speed and size and shape of the vessel being analyzed. A set of step-by-step simulation operations is embedded into the simulation file to make these decisions automatically. The operations will determine all the meshing sizes, places for refinement, where to apply the boundary conditions and all the relevant solver settings and stopping criteria. This process is so powerful, you won’t even notice it taking place and you press a button to mesh AND run the simulation; and these two steps are seamlessly linked together. What is perhaps even more powerful, is this automated process then allows for multi-mesh sequencing (MMS). This method automatically refines the mesh, maps the previous solution to the mesh and continues the run. This process is repeated systematically until the final mesh discretization is reached. On average, this process improves the run time by a factor of 4 compared to running the model only on the final grid! Illustration of MMS sequencing grid refinement. What about post-processing? Reporting for key metrics such as resistance, trim, heave or shaft power can be set up in advance. Field data such as friction coefficients, pressure coefficients and generated wave elevation can be displayed in pre-defined scenes and exported to PowerPoint using a macro. Key information, such as local wake fraction can be exported to CSV data automatically to be read into other software packages for further analysis, if required. Layout view of various automated post-processing visualizations. Got it – faster simulation runs. What about a complete design sweep? The template is fully parametrized, meaning that any type of design exploration study can be implemented using the design manager tool in Simcenter STAR-CCM+ . You can automatically explore the design space; generate resistance and powering vs speed curves, explore effects on mass, or analyze a range of possible hull forms and more! Want to understand how CFD simulation can transform your maritime projects, reducing time and effort, and still ensure compliance with environmental goals? Schedule a meeting with CAEXPERTS and find out how to bring your naval engineering to a new level with Simcenter STAR-CCM+ . WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

The role of Whole Engine Model (WEM) Predicting clearances in gas turbines is a complex and interdisciplinary task, involving transient thermal-mechanical modeling of all components within an assembly across various operational scenarios. This process, known as a Whole Engine Model (WEM), is crucial in the aero and power generation gas turbine industries. Understanding how clearances in gas turbines change over time is crucial for manufacturers. This knowledge allows them to optimize engine performance and efficiency while minimizing the risk of rubbing and other damage. The Whole Engine Model (WEM) isn’t just for assessing the steady-state condition of a gas turbine when it’s fully heated. It also helps analyze transient operations, where thermal gradients, temperature differences, and rotational loads significantly affect the gap between rotating and stationary parts. Clearances in gas turbines: understanding metal expansion Most of you probably remember physics classes and the experiments with metal utilizing the ball and ring expansion and the bimetallic strip examples, figure 1. Metal expands when heated and different metals expand at different rates. In a gas turbine, and in particular heavy duty gas turbines, there is a lot of metal that is heated at different rates and amounts resulting in axial and radial expansion over the course of the operational cycle. Figure 1. Metal expansion under thermal loads. Both images courtesy of Wikipedia Impact of clearances on performance and efficiency Gas Turbine OEMs are under tremendous pressure in a competitive market to deliver high performance and efficiency, both of which are intimately tied to the running clearances or gap between stationary and rotating parts. For example, 1mm of a clearance difference at turbine blade can have an impact of one megawatt (MW) which can power 650 houses in the United States (U.S.). That means a 500 MW gas turbine can power 325,000 homes, and just a 1 percent efficiency loss would mean powering 3,000 fewer homes. Alternatively, a study from Oak Ridge National Laboratory found that a 1 percent efficiency gain in a one-gigawatt (GW) power plant represents savings of 17,000 metric tons of carbon dioxide (CO2) per year, which is equivalent to taking over 3,500 internal combustion engine (ICE) vehicles off the road. Therefore, the clearances in gas turbines need to be as small as possible to ensure maximum power and efficiency. In the context of a rotor with a diameter of 1 to 2 meters and a length between 5 and 15 meters, the 2D WEM enables simulating and illustrating millimeter-level displacements to optimize performance and ensure component integrity. Evolution of the clearances in gas turbines On top of that, the OEMs need to ensure that no structural damage occurs as the clearances in gas turbine evolve during the start-up and shut-down of the engine. Figure 2. Example of the evolution of the clearances in gas turbine over a typical operational cycle. Graph is representative and movements are exaggerated for illustrative purposes Figure 2 is an illustrative example of what happens to one blade as an engine speeds and powers up. The cold build clearance, or gap that is present once the engine is assembled and cold, is arbitrarily set here at 2.5mm. The gap changes and we see initially it closing to a local minimum at start-up due to the centrifugal deformation and little to no thermal load on the stator side. A maximum clearance subsequently occurs as the gas turbine increases its load and more heat is added to the engine. The stator, consisting of relatively thinner parts warms up faster but as the rotor increases in temperature, we see an evolvement to a steady state running clearance. At shut down the load is switched off and we see a closure of the gap further as the stator starts to cool down before the rotor. The gap opens once the load and speed drops, and we see a gradual return to the cold build clearances as the engine cools down. Operational challenges and pinch points in gas turbines However, gas turbines have a wide range of operational scenarios, not just the idealized version representatively visualized in figure 2. Considering the current utilization of power plants as a back-up for green energy, the type’s of operation cycles the gas turbines have to endure are physically more demanding and characterized by frequent starts, stops and restarts. This has an influence not only on the lifetime of the parts, but also when and where local pinch points (tightest clearance or smallest gap) in a gas turbine occur. Figure 3 is an example of such a pinch point occurring at one blade location in the engine. One hour after shutdown, the engine is restarted and one can see how a new local minimum occurs as the rotor is still warm, the stator in comparison is cold and the centrifugal load is reapplied. Figure 3. Example of a 1 hour restart cycle. Graph is representative and movements are exaggerated for illustrative purposes Leveraging Simcenter 3D for clearance analysis Considering a typical industrial gas turbine, with approximately 15 compressor stages, and 4 turbine stages, the clearance engineer must carry out an analysis for each stage, and for all operational conditions, figure 4. This is where the WEM process within Simcenter 3D can significantly help the clearance analysis process. Reference points can be assigned within the model, for example, on the leading and trailing edges of the blades and vanes and the respective opposite locations on the stator and rotor. The transient results for the movement of these points can be read out from Simcenter 3D for further analysis in a tool like MS Excel or Matlab. This allows a clearance or mechanical engineer to build up a picture of the axisymmetric effects of the engine. Coupling these results to the non-axisysmmetric effects from other contributors (rotor-bending, casing deformation, etc), uncertainties and tolerances and the trusty engineer is able to build up a complete picture of the clearance behaviour of the gas turbine engine. Figure 4. Clearance analysis required for each blade and vane in a gas turbine. Inputs for the clearance stack-up from various contributors. This type of analysis can be used by the project team to define and optimize the performance of the engine, minimize the rubbing risks, and define the safety margins to be employed. Other decisions like abradables, squealer tips and ultimately the final manufactured and assembled state can be made with confidence using this a priori knowledge of the clearance behaviour of the gas turbine engine. Schedule a meeting with CAEXPERTS and discover how Simcenter 3D with the Whole Engine Model (WEM) can transform clearance analysis in gas turbines, optimizing performance, efficiency, and operational safety. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

In 1817, Karl Drais kicked things off by creating the Draisine, considered the starting point for the concept of the bicycle. This invention revolutionized the way people got around using only their own body strength. Since then, the evolution of bicycles has become heavily dependent on cutting-edge engineering. What characterizes today's great bicycles and components is the fact that they are the result of development processes based on rigorous simulations and testing. As in the automotive and aeronautical industries, virtual product development and advanced measurements have become fundamental to achieving superior performance, safety, and comfort in modern bicycles. Ride faster – Aerodynamics CFD and the power of GPUs Since the first Draisine a very obvious challenge in the improvement of bicycles was making them faster. Making a bicycle faster means less friction. And less friction means two things: First, lower mechanical friction in the drivetrain and between the wheels and the ground. And second, less air friction aka drag, which translates to better aerodynamics. And while tight pants can only take us so far, for high performance racing machines, CFD simulation had to conquer the world of bicycle development to make them cut through the air as efficient as possible. And to design a faster bike faster, GPU-native CFD solver technology arrived just in time as Trek Bicycle impressively demonstrated for at the Tour de France 2024. Ride more – Design space exploration But meanwhile companies like Trek not just run a one-off CFD simulation to see how a design works and then fiddle around a little with another design. No, they run automated design exploration workflows to explore the performance of hundreds of designs (that’s when cloud-hosted HPC and GPU-accelerated CFD comes in quite handy). To give you an idea of the level of care that goes into today’s optimization of race bike aerodynamics: Trek identified the optimum position of the water bottles to minimize drag, identified the optimum rider position under varying wind conditions and found the optimum drafting configuration four a team of four riders. On top these optimization studies are no longer only about one performance attribute. Taking it one step further, Trek looked at a multi-attribute optimization of aerodynamic performance and the weight of the frame. Ride Safer – Helmet impact testing SeIf you ride fast, you better ride safe. The good news is that helmets used today have improved significantly in terms of safety and comfort compared to those used in the past. This is thanks to continuous product improvement, novel materials and smart structural designs driven by homologation for certification and the customer desire for comfortable safety. Siemens professionals perform high-speed video analysis, laser head positioning, and wireless data acquisition to provide highly accurate and repeatable helmet testing. Experts can even assess the level of protection a safety helmet can offer against brain injuries caused by critical rotational accelerations. Ride cooler – Helmet aerodynamics and thermal comfort CFD Safety is one thing, but if you are on a long ride (thermal) comfort is another. But just like with vehicles, when it comes to helmets there is this constant fight between aerodynamics and cooling. Luckily, todays engineers have advanced CFD simulation at their fingertips. Below are just a few examples of how fluid dynamics combined with heat transfer enable insights to find optimum solutions in the trade-off between thermal comfort and low drag of helmets. Ride further – System simulation for ultra-lightweight hydraulic transmission Whatever your personal feelings, the success of e-bikes is actually great news. For human beings, for this planet. A lot of people who would probably have never taken the bike, would have taken their car even for short distances, would have claimed they would have been sweated all-wet, had it not been for the development of the electric bicycle to change their minds. Above all the success of the e-bike was driven by innovation in battery technology. But also the electric motor design and even rethinking transmission systems made the e-bike an interesting playground for engineers. Ride quieter – Harnessing Simcenter for testing e-bike acoustics When it comes to e-bikes range is by far not the only thing in scope of engineers. Noise is equally. And as you cannot just make an e-bike dead silent it’s about creating acoustic comfort for the rider. And guess what, leading companies are heavily investing into these acoustic experiences. Take Trek Bicycles again, they are pioneering the concept of putting sound quality on the e-mountain bike metric map as it continues to be a hot topic in the industry. Likewise, MAHLE, who use Simcenter testing solutions to streamline e-bike drivetrain noise-vibration-harshness (NVH) analysis and for end-of-line testing. Ride smoother- The grandfather of all comfort Speaking of vibration, we can't help but mention this: It is probably the most underrated part on almost any modern bicycle. Just a little bit more than an inch long, this tiny little piece has saved billions of bicyler’s a#*s over the last 130 years. If it wasn’t for Scotsman Dunlop with his tiny little invention we would probably still be riding harsh full rubber tires. But thanks to the invention of the Dunlop bicycle valve from 1891 (US patent US455899 ), we can now happily inflate our tires and enjoy the pleasure of air damping. Now, if you claim to be a bike enthusiast, can you explain how such a valve actually works? No? Well, here’s a little Fluid-structure interaction simulation that may give you some insights you will remember the next time you have to pump it… Ride braver – Simcenter for a safe low-drag monocoque wheel Ok let’s shift gears again. And in cycling shifting gears always means let’s talk Carbon. In the competitive world of cycling, innovative engineering plays a crucial role. At the edge of technology marginal gains can be the difference between winning and losing. That’s the playing field for Radiate Engineering & Design AG. Radiate’s collaboration with Scott Sports led to developing the Syncros Capital SL, a wheelset that makes every bike enthusiast go quiet in awe. The standout feature: “The Capital SL is a one-piece, or what we call monocoque, construction, meaning the rim and the spokes are fused together into a single piece,” says Frederic Poppenhäger, a partner at Radiate. This design enhances structural integrity and reduces rotational inertia, leading to better power transmission and substantial speed gains for riders. By running simulations, Radiate evaluated various rim shapes and geometries, optimizing for aerodynamics and structural performance without the immediate need for physical models. While Radiate reduced drag by 7 percent they also improved rider confidence and safety by ensuring greater stability in crosswinds. To enable a braver ride. Ride together Schedule a meeting with CAEXPERTS and discover how Simcenter solutions can transform bicycle development—from aerodynamics to comfort, safety to performance—taking your engineering to a new level of innovation. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Cryogenic storage and distribution — handling substances at extremely low temperatures — might sound like something out of science fiction, yet it plays a crucial role in many engineering applications. From aerospace rockets to medical preservation, superconducting technologies or liquified natural gas (LNG) in ships, cryogenic systems are everywhere. But how do we design these systems to be efficient with a maximum level of safety? Using thermofluid system simulation can help do just that . In this blog post, we will explore how Simcenter Amesim , a top-tier simulation platform, empowers engineers and enthusiasts alike to model and optimize cryogenic storage systems efficiently using a simulation example inspired by a very well-known NASA experiment of cryogenic tank self-pressurization. Challenges of cryogenic storage Cryogenic storage involves storing fluids (hydrogen, nitrogen, natural gas) that are gaseous in ambient conditions. Decreasing their temperature below a certain point changes them into very cold liquids, critical in certain applications. The storage of these cryogenic liquids requires specialized insulated containers, like dewars or cryogenic tanks, to maintain low temperatures, prevent heat transfer, and ensure safety. But with low temperatures come unique challenges such as thermal management, pressure buildup and boil-off. Dealing with challenges requires advanced simulation capabilities to frontload any safety issues that might occur during the product life cycle. Simulating cryogenic storage in Simcenter Amesim Simcenter Amesim offers comprehensive tools for modeling and analyzing cryogenic storage systems. It comes with a set of key components such as cryogenic tanks that enable simulations that closely represent real-world conditions, including factors like gas-liquid interaction and thermal exchange. Core capabilities: Heat and mass exchange at the liquid/gas interface Simulate the liquid and gas phases within a storage tank Model scenarios including filling and emptying, self-pressurization and boil-off Why Use Simcenter Amesim for cryogenic storage? Simcenter Amesim comes with a library of fuel cell and cryogenic fluid storage libraries. Compatibility between these multiple thermofluid libraries and the possibility to couple with detailed lumped thermal model facilitates system-wide integration, enhancing a holistic evaluation in engineering projects. With these capabilities, from the beginning of the design phase, the user can easily estimate the effect in terms of pressure buildup and temperature caused by possible heat ingress in a cryogenic tank. Thanks to the addition of a film layer at the free surface of the tank, pressures and temperatures are captured with increased precision to allow engineers to better size insulation systems. This results in 3-node cryogenic tank model – bulk, film and ullage. Three-node cryogenic tank in Simcenter Amesim As for the tank geometry, it can be of any 3D shape – thanks to the Simcenter Amesim tank CAD mapping tool, we can generate the tank liquid height as a function of volume. A practical example: NASA’s experiments of self-pressurization in a liquid hydrogen tank With the capabilities explained above, a Simcenter Amesim model can be built focusing on boil-off and self-pressurization in a cryogenic tank. This demo is based on experiments at NASA’s Lewis Research Center , exploring liquid hydrogen (LH₂) storage in a 4.89 m³ spherical tank subject to a constant heat ingress —findings published by Hasan et al . As for the setup of the experiments, the LH₂ tank was enclosed in a cylindrical cryoshroud. The shroud may be cooled with liquid nitrogen or heated above ambient with electrical resistances to maintain a certain temperature around the tank that we call Tamb in this demo. Understanding boil-off rates Three boil-off test cases were performed at 83 K, 294 K, and 350 K of ambient temperature. To cool the upper section, the tank is filled with LH₂ to 95 % capacity. The vent pressure is then gradually reduced to the backpressure control system’s operating pressure of 117 kPa. The boil-off rate is monitored until it stabilizes. The Simcenter Amesim model used to perform this test case is shown below: Hydrogen boil-off model In the above model, GH2 is vented from the tank through a relief valve. Heat is also transferred from the exterior to the vapor on one side, and from the exterior to the liquid on the other side. For each case, a heat transfer coefficient is set using a variable thermal conductance to fit the average heat flux values absorbed, as shown in the table below. The thermal conductances are piloted with the wet and dry areas to compute the heat flows. ] Experiment Ambient temperature [K] Heat flux [W/m²] Exp 1 83 0,35 Exp 2 294 2.0 Exp 3 350 3,5 The final values of the boil-off rates (expressed in SCMH) from the model, after 50 h of simulation, are pretty close, as shown below next to the steady-state boil off rates of the experiment. Boil-off rates (simulation vs experiments) Exploring self-pressurization dynamics Self-pressurization in a cryogenic tank is basically when you leave the tank sitting in “hotter” environment, LH₂ will evaporate and increase the tank pressure – which may cause a safety issue at some point, so it’s important to assess this pressure buildup in the cryogenic tank. Self-pressurization tests were conducted at different ambient temperatures: 83 K, 294 K and 350 K. The tank initial fill level was 84%, the initial pressure was 103 kPa. The goal here is to set up the model to match the time-series values of tank pressure and check whether time-series values of temperatures are within the correct ranges. Self-pressurization model In the model above, there’s no venting. LH₂ is stored and the variable thermal conductances are set to match the average heat fluxes exchanged with the ambient. Just like the boil-off tests, there are 3 self-pressurization tests, lasting respectively 20 h, 18 h and 14 h. Pressure values are shown below. Pressure values (simulation vs experiments) It can be observed above that for the pressurization tests, the pressures computed by the model exhibit a similar pattern to the tests. They deviate from the experiment by an absolute maximum ranging from 1.5 to 6 %. Conclusion As seen in the current demonstration, with a cryogenic tank model divided into three nodes — ullage, film and bulk, we can capture important phenomena such as boil-off rate and self-pressurization. For enhanced temperature predictions in ullage where different temperature layers can exist, further discretization of the upper part could prove beneficial. Cryogenic storage is a pivotal technology shaping the future of industries relying on hydrogen and other low-temperature applications. Simcenter Amesim offers not just a platform, but a comprehensive toolset to visualize, model, and optimize these systems. By leveraging its capabilities, you can significantly advance your understanding and design of cryogenic storage solutions. Ensure your cryogenic projects are more efficient and safer with the power of Simcenter Amesim . Schedule a meeting with CAEXPERTS now and discover how our expertise can help your team model, predict, and optimize cryogenic storage systems with precision and confidence. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Electromagnetic (EMAG) and electrical engineers feel at home with EMAG's simplified Computer-Aided Engineering (CAE) packages, such as Simcenter SPEED , Simcenter E-Machine Design , and Simcenter MAGNET . Computer-Aided Design (CAD) colleagues generate a defective geometry consisting only of active EMAG parts. If there are CAD problems, such as extracting geometry from an assembly or mesh errors, the design is sent back to the CAD designers, and the clock starts ticking. Can we continue like this? With the race to electrification, there is a need for more cooperation across different teams. Product designers give us “the look and the feel” of the product – the skin. CAD designers distribute real estate. CAE engineers then work within constraints to add the physics needed to optimize the product. Spending some time on communication and alignment between teams is seen as an inherent part of the process “it’s just how these things work in big companies,” is a common aggrievement. Sub-optimal processes may have been fine when electromagnetic components were an accessory to your product. However, nowadays, an electric drive is at the core of many products, and wasting time in iterations with other departments can double your development cycles and skyrocket R&D costs. Furthermore, the high competition and the lack of engineering resources make it unsustainable to keep wasting time on meaningless iterations. What about upstream CAD changes that affect the EMAG bodies? It is not unusual for an EMAG engineer working on an e-powertrain to receive a detailed CAD neutral file. Often these files, although consisting of a housing and transmission assembly, are non-associative (we will discuss what non-associative means later). The files will be missing the rotor, stator core, or both and are more visual than functional for CAE. This sets off a back and forth between the EMAG and CAD teams, who now need to clean up and simplify the geometry and verify the materials. Only after this clean-up process can an electromagnetics engineer start their real work of searching for a concept and validating the proposed design. Due to the current soaring interest rates and higher costs, Project Managers in the EV (Electric Vehicles) space are under elevated pressure since 2018, to hit vehicle delivery timelines at elevated costs if left unchecked. As a result, you may hear yours asking your team to shed weight and volume from components. As an EMAG engineer, you will find yourself crafting a way to hustle the CAD designers of their time. Rest assured; you aren’t the only one. CAD designers must reflect the new CAD updates and help CAE teams update their geometries. As you can imagine, the emails, calls, and meetings to make this happen are extensive, and the impact on the timeline to market is enormous. How to save time? A system that allows the electromagnetics engineer to automatically update their models to the latest CAD design, with minimal input from the designer, will increase productivity. You can see an example of a possible process in this short video. The video shows how to update an extracted 2D e-motor geometry from a vehicle assembly due to upstream changes to the stator outer diameter. The video begins by focusing on the electric powertrain and gradually isolates the electric motor from the rest of the assembly. Next, not only are the active electromagnetic bodies captured, but they are also associated with the assembly. Next, everything needed for the simulation is added: the air regions and remesh, the 2D geometries of the rotor, and the finite element mesh to complete it. Finally, an upstream change to the outer diameter of the stator is triggered, which you can see in the updates to the mesh. The CAD connection in the Simcenter environment allows you to focus on your work—electromagnetic analysis. Extracting and associative 2D e-motor geometry from a vehicle assembly Now let’s turn our attention to the global push for lightweighting and integrating e-powertrains. There is a need to control noise levels and get rid of the heat in ever smaller electric –motor sizes. These need more complex liquid cooling, and designs that ensure the magnets do not fly off, all of which will change the electromagnetic’s structure. How do you update the electromagnetics model with multi-physics-driven geometric changes? In a traditional workflow, the non-associated model updates made by other groups would require the EMAG engineer to repeat the workflow hustle with the design team and all its associated emails and calls. You will feel damned if you have no associativity! On top of that, the problem only gets worse by increasing the number of physics domains under consideration. E-powertrains undergo an entire range of multi-physics simulation iterations before the engineers settle on a design. These horizontal changes are mutually driven by electromagnetics, thermal, structural, and noise and vibrations (NVH) analysis. With engineers from each department working simultaneously with a non-associative tool, the possibility of an engineer working on an out-of-date model instead of the current best design is remarkably high. How do you keep track of the updated CAE geometry? Luckily, there is a reference – though subtle in a siloed environment. The design CAD is the only common link between physics, and the product requirements, such as cost, which is pegged to the volume and mass. As illustrated in Figure 1, if the design CAD is seamlessly updating, it means all the CAD and all CAE engineers are accessing the same geometry at any one time. Obviously, the updating and access are controlled in a managed environment. This means you won’t have stale results in your project meetings. For electromagnetics, geometry will reflect the recently approved upstream and horizontal changes. Figure 1 - An updated design CAD captures geometric changes driven by both upstream product requirements, and horizontal physics interactions For example, Figure 2 shows the evolution of EMAG stator parts from an efficient design based solely on electromagnetic performance to one that accommodates cooling channels in the coils. Note that there is a reduction in torque because the volume is now limited. This is because the first design contains more steel and can therefore carry more magnetic flux. As you can see in the video, you can associate the electromagnetic’s active parts with the CAD design. After a parametric study to choose the final channel width, this geometric change was introduced into CAD design. And because the geometry used for the electromagnetics simulation is associative, it was just a matter of re-running the solve with the changes Figure 2 - The effect of introducing cooling channels between coils on EMAG performance What does this really mean? In a multi-physics design process like the one required for an e-machine, having a streamlined workflow can save hundreds of engineering hours. You only spend the initial setup effort once! The rest is just on iterating the design driven by both product requirements and physics changes. The associated workflow helps overcome the compartmentalized approach and single physics bias in development. An efficient electromagnetic device may not be structurally sound or even manufacturable. By capturing what is possible through geometric constraints and physical interactions, a more realistic design is achieved. At the same time, this helps manage our own internal bias toward physics and work together toward a good balance of performance. What you think is appropriate for the physics of the design may not necessarily be good for the overall performance of the product. An integrated design process allows you to make these iterations and converge much faster! In summary, the associated process discussed in this blog reduces time wasted so products can be developed quicker. This process helps engineers and designers settle on a concept quicker by ensuring everyone is working on the most up-to-date model. While ensuring that disadvantages such as bias towards particular physics models or siloed teams do not occur. Build your next e-motor quicker, better, and in terms of engineering hours, cheaper. Are you ready to reduce rework, accelerate your development, and integrate CAD and CAE teams more efficiently? Schedule a meeting with CAEXPERTS and discover how to optimize your electric motor design with associative and multiphysics workflows—faster, more collaborative, and more cost-effective. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Artificial intelligence (AI) has become a ubiquitous force in the engineering world, radically transforming how companies operate and how employees work. As we move further into 2025, the influence of AI will only continue to grow, ushering in a new era of increased efficiency, productivity, and innovation. Nvidia CEO Jensen Huang has predicted recently that computing power driving advances in generative AI is expected to improve “a millionfold” over the next 10 years, driven by a “fourfold” increase in computing power annually. Driving this rapid advancement is the exponential growth in computing power over the past decade. Today, the world’s leading supercomputers are capable of performing over 1 quintillion (1,000,000,000,000,000,000) calculations per second – a staggering increase from just a decade ago. This massive boost in raw computing power has enabled the training of ever-larger and more complex AI models, allowing them to tackle increasingly sophisticated tasks. Powering this AI revolution are not just traditional CPUs, but also specialized hardware like graphics processing units (GPUs) and application-specific integrated circuits (ASICs) designed explicitly for machine learning workloads. These chips can perform the parallel processing required for training and running AI models much more efficiently than general-purpose CPUs. In addition to advancements in hardware, the rise of cloud-based platforms, such as Simcenter X, has also played a pivotal role in democratizing access to immense AI computing power. Researchers and businesses can now rent GPU clusters and supercomputing resources on-demand from cloud providers, allowing them to train large-scale models without the need for massive upfront investments. Optimizing business processes with AI Simcenter Engineering Services has demonstrated that the influence of AI extends far beyond digital assistants and personal productivity. Businesses are increasingly deploying AI-powered systems to optimize core operations and decision-making. AI algorithms can analyze massive amounts of data to uncover insights that inform everything from supply chain management to marketing strategies to financial forecasting. Engineering design decisions based on heat transfer and computational fluid dynamics (CFD) simulations are no exception. AI algorithms now support the entire simulation workflow chain, from data management and part shape recognition to real-time flow and thermal results predictions and optimization. AI/ML in CFD The synergy of CFD and machine learning promises to deliver faster solutions to well-known complex industrial fluid flow and heat transfer problems – most notably, design optimization of transient problems, such as vehicle aerodynamics and passenger cabin comfort. Vehicle aerodynamics Transient aerodynamic simulations, especially those involving complex geometries, require significant computational resources in terms of processing power and memory. Running these simulations at a large scale can be computationally intensive and time-consuming. Additionally, transient simulations often require very small time increments to capture dynamic air behavior accurately. This can lead to long simulation times, particularly for problems with a wide range of time scales. Additionally, the large amount of data generated by large-scale transient CFD simulations can be difficult to manage, store, and analyze. AI algorithms address this problem by providing inferred CFD results, including pressure distributions and air velocities, enabling the calculation of drag and lift coefficients across various design parameters. The AI models are trained on a set of simulation data, which consists of point clouds extracted from simulations at both the surface and volume levels, along with scalar values such as pressure, wall shear stress, and the three velocity components. To develop accurate AI models, it is crucial to execute the training process efficiently. Simcenter Engineering Services addresses this need by employing a hybrid technique designed to cover the vast design space effectively within a set number of evaluations. Initially, a collection of design points is established. These can either be provided by the user as seeds or generated through advanced techniques like Latin hypercube sampling methods. Once established, these initial points serve as inputs to an adaptive sampling study. This study is pivotal as it intelligently selects and populates additional design points aligned with user-defined objectives. Such objectives may include the exploration of regions where there are considerable shifts in response or areas where global and local gradient variations occur. The AI model itself is principally trained using spatial data from the surface and volume mesh cells. In addition, it incorporates feature-defining scalars, such as surface normals, alongside the scalars intended for prediction by the AI model. This comprehensive approach ensures that the model can accurately predict complex engineering scalars spanning over varied design changes. The image below demonstrates an example of the training and inference stages of an AI model, showcasing its application in utilizing spatial coordinates, surface normals, and wall distance to predict engineering scalars like pressure and velocity vectors. Training workflow Inference workflow Passenger comfort Running transient CFD simulations for a vehicle passenger comfort analysis is often necessary and comes with many unique challenges. The vehicle cabins are known to have intricate geometries with many small features, such as vents, seats, and other interior components that must be captured in detail using high-resolution meshes. For unsteady flow, this meshing requirement can be quite computationally demanding. Additionally, the cabin passengers must be modeled in good detail, with separate output required for the manikin head, neck, torso, hands and feet. This level of refinement is necessary because human beings feel most comfortable within a relatively narrow range of ambient temperatures. Even small deviations outside this range can start to feel uncomfortable. Adding to the complexity is the fact that manikin surface temperatures can be influenced by very small deviations of the environment conditions, such as air flow, temperature, humidity, solar radiation, vehicle speed and acceleration. Recognizing these challenges, Simcenter Engineering Services is actively exploring reduced order models and AI/ML-based solutions specifically tailored to address challenges in passenger comfort analysis. One area of focus involves the use of reduced order models (ROM), incorporating both static ROMs and proper orthogonal decomposition (POD) interpolation. Static ROMs are used to predict key scalar metrics, like temperature or velocity, in a text-based format that provides essential quantitative insights. On the other hand, POD interpolation is employed to predict spatially varying 2D scalar fields. This approach enables the determination of key scalar metrics, such as cabin temperature and velocities, in two dimensions based on input parameters like solar loads, the sun’s relative position, and flow characteristics. Both reduced order models can be generated using simulation (and test) data for training. The combination of these two methods provides a simplified yet effective means to analyze complex data, ensuring that critical passenger comfort parameters are accurately assessed and optimized. These and other types of reduced order models can be generated using Simcenter Reduced Order Modeling tool available in the Simcenter portfolio. Complex 3D modeling AI algorithms can enhance and accelerate the model-building phase of complex 3D thermal simulations. In vehicle thermal and energy management (VTM/VEM), accurate part-to-part contact through CAD imprinting is crucial to capturing heat conduction paths throughout the vehicle assembly. Achieving high-fidelity results often requires manual review and improvement of hundreds of poor-quality imprints, leading to extended project turnaround times. To improve efficiency and reduce turnaround times, Simcenter Engineering Services is developing custom automation and “smart logic” to support the CAD imprint and review process. This leverages AI/ML-based procedures within Simcenter STAR-CCM+ . Various “modalities” can be employed for inference, as illustrated below. Once the AI model is trained, contact validity can be inferred through a custom frontend plugin using the selected modality. This approach integrates both tabular and image-based data training to derive inferences from the ML model, expediting and simplifying the modeling process. For straightforward contact validity predictions, a simple classifier model using features from a text-based file is sufficient. However, when dealing with complex contact patches where additional context, such as the identity of contacting parts, is necessary, an image-based model offers greater advantages. This model can analyze visual data to assess the spatial and structural relationships between components, providing a more nuanced understanding of contact validity. The inference pipeline for both tabular and image-based model are shown below: Inference pipeline for tabular-based model Inference pipeline for image-based model This innovative approach not only speeds up the modeling process but also enhances the accuracy and reliability of thermal simulations, ultimately leading to improved vehicle design. The path forward There’s no doubt that AI is reshaping the engineering landscape in profound ways. Savvy leaders and innovators are embracing AI as a transformative opportunity, strategically deploying AI to augment and empower their workforce. As an industry trailblazer Simcenter Engineering Services is leading by example, empowering employees and customers with AI-driven tools and capabilities, enabling the engineering community with the ability to unlock unprecedented levels of productivity, creativity, and business success. Discover how CAEXPERTS can transform your engineering projects with AI and CFD, accelerating complex simulations and optimizing critical processes. Schedule a meeting with our experts now and take the next step toward innovation and maximum efficiency. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

Smoother lubrication with oil injection Oil injection offers several advantages over traditional oil bath lubrication in gearboxes. This technique allows for a significant reduction in oil consumption, while providing greater precision compared to oil bath lubrication. The process enables precise control over lubricant supply, ensuring optimal lubrication at critical points. The localized application of oil helps minimize waste, resulting in lower overall consumption. However, lubrication by injection requires a higher level of knowledge of the flow pattern and system dynamics. Moreover, it usually demands higher initial investment compared to oil bath systems. Therefore, detailed engineering insights in the early design phase of a gearbox are critical to realize optimum injector positioning and injection strategies. A viable tool to cope with this challenge early in the design of a gearbox lubrication system is Smoothed-Particle Hydrodynamics (SPH). Shifting gears with SPH fluid injection Since version 2402, Simcenter STAR-CCM+ has incorporated SPH technology into an integrated CFD environment. In previous versions, the Smooth Particle Hydrodynamics (SPH) solver already allowed simulations of powertrain lubrication in oil bath configurations. With version 2406, the functionality has been expanded to cover injection scenarios through the inclusion of specific inputs for SPH. This makes it possible to simulate oil injection with speed or mass flow imposition, as well as to define constant or time-varying distributions, such as in the lubrication system start-up process. This makes it possible to visualize the behavior of the lubricant inside the gearbox with SPH and inlet boundary conditions. This helps to validate the direction of the oil in the gear elements, ensuring efficiency and reliability in lubrication. Stay integrated with the power of Simcenter STAR-CCM+ as a platform Thanks to the integration of SPH into our flagship Multiphysics platform, you can now harness Simcenter STAR-CCM+’s full suite of quantitative data analytics capabilities in conjunction with smoothed-particle hydrodynamics. For gearboxes, you can e.g. monitor churning losses by looking at the forces and torque evolutions throughout your simulation. The torque report and graph do not depend on the surface mesh resolution. Even if you have a coarse representation of the surface mesh of your geometries, it will no longer affect the accuracy of the torque evolution. This ensures accurate results regardless of the details of the surface mesh. More applications with rotating injectors In the example above, you saw a static SPH fluid injection. Static inputs are commonly used when the flow entering the domain is not influenced by rotation. With Simcenter STAR-CCM+ , it is also possible to set up a rotating boundary condition. In injection lubrication systems, rotary injectors are commonly used to supply oil lubricant to moving parts such as gears or bearings. The rotating inlet condition ensures that the lubricant flow enters the domain with the desired rotational movement. Proper modeling of this behavior is crucial for accurate simulations of powertrain lubrication performance. This animation shows a simplified lubrication system in electronic machines. Using SPH for this application makes it easy to monitor where the lubricating oil goes with a simple workflow, compared to finite volume models. Other applications are also possible thanks to the addition of inlet boundary conditions. You can start modeling water runoff or oil leakage applications in vehicles by observing where the liquid (water or oil) enters the vehicle. You can observe how the liquid spreads to different parts of the vehicle, where it accumulates, and how it interacts with other components. To mesh or not to mesh This question has remained relevant since the introduction of SPH: to mesh or not to mesh. As in previous versions, SPH offers distinct advantages for specific applications. Traditional CFD simulations often require CAD preparation and volumetric mesh generation, which can be time-consuming for systems such as gearboxes. SPH eliminates this step by operating directly with particles, providing a more fluid experience. The combination of intuitive workflows, CAD integration, and robustness in handling movements sets SPH apart in Simcenter STAR-CCM+ . In addition, an industrial gearbox can be simulated with SPH in about 12 minutes. Thus, for initial design screening, the SPH method is ideal for use with Design Manager for design exploration, quickly and easily eliminating low-performance designs. If your goal is to visualize the oil distribution inside the gearbox and monitor agitation losses, SPH is the best candidate for this application. However, if air significantly affects the oil, or if you need to model spray inlets or observe heat transfer, or if you need to model mist (gas and liquid mixture) or model phase change or electromagnetism or other complex physics, it is recommended to use finite volume multiphase models, which are more suitable for greater accuracy. The SPH and finite volume methods complement each other and always depend on the scope of your applications and the level of fidelity required to make the appropriate choice of model to use. Explore more results from the gearbox lubrication SPH simulation interactively in your browser Surface tension modeling for SPH The surface tension model was introduced with the aim of increasing accuracy in highly dynamic free flow simulations. This functionality allows the behavior of droplets and their interaction with surfaces to be represented. Contact angles can be defined for each solid contour, enabling simulations of hydrophilic or hydrophobic surfaces. Ensure maximum efficiency in the lubrication of your gearboxes from the earliest stages of design! Schedule a meeting with CAEXPERTS and discover how to apply SPH in Simcenter STAR-CCM+ to accurately simulate oil injections, reduce losses, and accelerate engineering decisions with confidence. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

In the context of vehicle electrification, the automotive industry is taking several key steps to ensure the safety and durability of vehicle batteries. Rigorous testing protocols are in place to evaluate battery performance, thermal management and response to abuse conditions like overcharging, short circuits, and crashes. Battery ageing is being looked at intensively, and even more is battery thermal runaway, for very good reasons. The key to properly managing these requirements is front-loading the design. This means using Digital Twins to predict these effects early on in the design process, thereby minimizing the cost of physical testing and troubleshooting. But what about structural durability? Typically bolted below the body, the battery experiences the same loads as the body when the vehicle rolls over potholes, cobblestones and speed bumps. This repetitive loading will lead to accumulated damage in the structure of the battery frame and cause premature failure and high cost if not engineered appropriately. But can more be done? Can the same front-loading battery life strategy be applied using Digital Twins, allowing us to predict and understand the structure's behavior from the outset? SIEMENS simulation expert Dirk von Werne investigated this and developed a process that allows this to be achieved. This process utilizes SIEMENS testing solutions to identify appropriate load conditions and numerical simulation expertise. Besides front-loading, it allows us to understand the effect of the body to which the battery is attached. Indeed, the battery is reinforcing and interacting with the body structure. As they form an integral structure together, the loads that the battery experiences are much dependent on the stiffness and eigenmodes of the full system. This leads to complex load scenarios and responses during driving, that are not easy to replicate in component testing. The ability to include these effects in simulation allows us to engineer a better battery considering the coupling to the body, and to develop an improved accelerated test scenario for component testing. Finally, this approach also allows us to minimize the mass of the battery pack without compromising durability. Indeed, the battery is one of the heaviest (and most expensive) components of an electric vehicle. Duty cycle and loads As a starting point, you have to define the mission profile that you design your vehicle for. Standards are available to support this, but they only represent a typical usage and might not fit with a specific vehicle under development, a given target market, etc. Therefore, collecting data in relevant test conditions on a relevant vehicle should be considered whenever possible. This can be done, for instance, on a predecessor vehicle. Many scenarios can be considered: how many kilometers one will drive on the highway, in the city, or on a smaller road with cobblestones? What speed ranges to take into account? Which driver profile: comfort driver or sportive driver? All in all, finding the proper loading conditions for a vehicle under development is far from an easy task! Simcenter Testlab Mission Synthesis is a software that condenses the loading scenarios from years of operation to an equivalent accelerated test cycle. This allows you to understand the “damage potential” in many different conditions and synthesize your “test profile” much better. This can then be used for validation testing and also in a simulation context, where these loads are used as input to the Digital Twin of the structure. Mission synthesis generates a test profile Knowing the stress In the world of automotive durability, simulation processes are very well established based on decades of vehicle engineering. Using predecessor knuckle forces or actual road surface profiles as an input, you can simulate the vehicle behavior in a multi-body simulation environment, such as Simcenter 3D Motion , deriving stresses and interface forces at the battery attachment. Full-vehicle model to predict component loads A Finite Element simulation in Simcenter 3D will assess the damage and life prediction of the battery frame based on the loads at the attachment points, modal stresses to be used in time or frequency domain response analysis, and the superposition of load cases. Damage hotspots will be post-processed and their local data can be analyzed in detail. Yes, but… the battery is a very complex structure, including special materials, many connections, and some non-linearity. The battery’s structural coupling with the body structure is strong. For those reasons, applying the “classical” approach described above is not straightforward. This is where Simcenter Engineering comes in, with its experience in both testing and simulation. As Dirk explains, “You need to understand how the structure really behaves when subject to vibration loads. You can expect that a battery has a non-linear behavior, and need to find the best approach to capture it in your Finite Element model. Thanks to our large experience in combining test and simulation, we can support our customers in defining proper assumptions. Modeling guidelines can be defined based on the correlation of the physical battery with the simulation model, and adapting the model until the match is good enough. In the end, you get a Digital Twin that provides a reasonable representation of the actual battery”. The overall process for battery structural durability assessment is shown below: Workflow for battery durability analysis An optimal solution As mentioned before, there is tremendous pressure in reducing vehicle weight, in order to maximize the range. Numerical optimization enables to do this. To allow for an optimization, the durability prediction has to be fast enough to be looped in an automatic optimization process. This requires to balance the model size with the accuracy. One way to do this is to represent the Body by means of a reduced representation as a superelement. The optimization process shown below is driven by HEEDS , part of the Simcenter portfolio. A typical objective of the optimization is to reach the lightest possible battery frame while ensuring good durability performance. Typical optimization parameters include the cross sections and material thickness of the battery frame, as well as the bolt positions of the attachment to the body. All this becomes possible by parameterizing the CAD geometry in NX and building up automatic meshing, analysis and post-processing. The optimization in HEEDS is enhanced using AI. Optimizing the structure to reduce weight Conclusion In this blog post, a process for CAE-based durability analysis of an EV battery was presented. Durability simulation allows frontloading the mechanical design, optimizing the structure and developing a valid accelerated scenario for component testing. What’s in it for you? Obtain improved design earlier thanks to a CAE-based process Avoid conservative approach and reduce mass Improve and shorten battery testing Anticipate structural challenges and optimize the performance of your electric vehicle battery with a Digital Twins-based approach. Schedule a meeting with CAEXPERTS and learn how to apply advanced simulation and intelligent testing to ensure durability, reduce mass, and accelerate development with greater accuracy and lower costs. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br

For a new technology to succeed it must arrive at the right time, provide a solution to a real-world problem, and solve that problem better than any other competing technology out there. That might have seemed like the case when the first electric cars started to appear in the 1880s, at last a less polluting vehicle after the coal guzzling steam cars and carrot guzzling horse drawn carriages that went before. Cities were being electrified and other electric vehicles such as trams became common. In this environment, the EV started to gain traction at the end of the 19th century (if you pardon the pun), although adoption was limited to US and European cities where electrical infrastructure existed. It was not yet to be the era of the EV however, as the initial rapid growth of the technology hit a wall in the 1920s overtaken by the Internal Combustion Engine and cheap and abundantly available oil. The electric vehicle was relegated to be forgotten in dusty archives, where it lay quietly forgotten for most of the next century… There it remained, until the elephant in the room that was the climate crisis could no longer be ignored and at last the time was right for the EV. Of course, electric vehicles are not without their challenges – mainly related to the battery and motor. In the world of Engineering Simulation, battery thermal runaway and e-motor cooling have become very hot topics (literally!). These are simulations that would have been unthinkable just a few years ago due to their complexity both in terms of geometry and Multiphysics. Consider e-motor cooling for example, it is highly complex due to the range of scales involved. Cooling jets of oil breakup into ballistic droplets and then ever smaller droplets due to the high rotational speeds and interaction with the intricate end-windings. Clouds of droplets of tens of microns or less form mixtures and even foam. Thin films of oil form on many surfaces. In other places oil pools. Oil flows through tiny gaps going where it should not. All of this happening alongside heat transfer and electromagnetics. Hybrid Multiphase Comes of Age These complex applications pose new demands of simulation, the tools of the past are no longer sufficient and the time for hybrid multiphase is here! Individual multiphase models like Volume of Fluid (VOF) and Mixture Multiphase (MMP) were designed for a single multiphase flow regime. The VOF method for example expects everything to be resolved and has no concept of a mixture, whereas Mixture Multiphase expects only mixture and no resolved free surfaces. Pick any one method for these new applications and it will be pushed beyond its assumptions and produce inaccurate results. For this reason, the move to more complex multiphase applications drives a need for hybrid multiphase. Adding Mixtures to Hybrid Multiphase In Simcenter STAR-CCM+ , Mixture Multiphase has been put at the heart of hybrid multiphase to enable applications such as e-motor cooling with three great new features: 1. Lagrangian to Mixture Multiphase sub-grid transition model The first feature is the ability to transition small Lagrangian droplets to a dispersed mixture phase. It can also be used for bubbles or solid particles, but focus on liquid droplet phases here. Lagrangian multiphase is well suited to modelling large, high momentum droplets that travel in a different direction to the surrounding gas, but when simulating the secondary break-up of droplets, you can end up with a large population of tiny droplets that follow the gas path. This is essentially a mixture and Lagrangian would not be a good model choice. With the new Lagrangian to Mixture Multiphase sub-grid phase interaction model, Lagrangian droplets can be transitioned to a mixture representation in Mixture Multiphase. This allows Lagrangian multiphase to be used only for high momentum ballistic droplets whilst the larger population of smaller droplets is transported as a mixture. This new functionality works with MMP-LSI – a model that combines the capabilities of VOF and Mixture Multiphase, using interface capturing locally where needed and allowing mixtures elsewhere. With this combination, we can now model the full cascade of multiphase scales from the resolved to the sub-grid. In the example below, we resolve the initial stages of breakup of a jet in crossflow using Large Scale Interface (LSI) modelling, before passing the resolved droplets over to Lagrangian where they continue to breakup into ever smaller droplets. Once the droplets are small enough and traveling with the surrounding gas they are passed to an Mixture Multiphase phase as a mixture. Transition is based on droplet size and Stokes number. 2. Population balance modeling for Mixture Multiphase Large Scale Interfaces (MMP-LSI) Now that we have a predictive pathway for the size of droplets produced by the breakup of large bodies of fluid, such as jets, we need a method to capture this valuable data and use it in simulations. Droplet size is an important variable in determining the transport of mixtures, and critical for physics such as evaporation or reactions. For that reason, support has been added for the S-Gamma population balance model with MMP-LSI. The S-Gamma model transports a distribution of sub-grid droplet sizes which it modifies every time Lagrangian droplets are transferred. In addition, S-Gamma models the effect of further breakup and coalescence. This is critical in applications such as e-motor cooling where jets of oil breakup into ever smaller droplets that may be 10s microns in size that are carried with the rapidly rotating air and even whipped into foam. We can see this model in action in the jet in crossflow example below where the predicted droplet size distribution from the jet breakup can be seen in terms of the Sauter Mean Diameter of the resultant sub-grid droplet population. 3. Lagrangian impingement into Mixture Multiphase continuous phase The third and final feature is another new phase interaction model – the ability of Lagrangian droplets to impinge into continuous phases in Mixture Multiphase. In hybrid multiphase simulations the same fluid can be represented by multiple different multiphase models and tracking oil droplets through liquid oil (for example) would make no sense. This new phase interaction allows Lagrangian droplets to impinge into larger bodies of fluid represented by MMP-LSI. Let’s see this in practice – below we have a fountain where the jet is initially resolved as a free surface (LSI) which then breaks up into Lagrangian droplets. The droplets then impinge into the free surface of the fountain’s reservoir, completing the cycle. Together, these capabilities allow you to simulate applications that contain mixtures or within the framework of hybrid multiphase. E-Machine cooling – A practical application of the new hybrid multiphase method Previously no toolset existed that could cover all of these multiphase regimes accurately in a single simulation in an affordable way. The marriage of these complementary technologies allows us to accurately simulate applications where before we had to suffer the inaccuracy caused by assumptions, or the expense of a high level of resolution. With hybrid multiphase we have affordable accuracy at our fingertips. The time is now right to model the complexity of electric vehicles (and other applications) with hybrid multiphase to gain the edge in performance and gain insights into your designs never before possible. See these tools being used to simulate oil cooling of an electric motor. Simulation of oil jet cooling in an electric motor becomes highly accurate yet affordable by using Hybrid Multiphase with Mixture Multiphase Large Scale Interface (MMP-LSI) including S-Gamma, Lagrangian Multiphase and Fluid Film Here, we have oil jets introduced from a rotating shaft, which has been solved with adaptive meshing and Large Scale Interfaces (LSI). The jets hit the end windings, cooling them and then breaking up into droplets. These droplets were tracked as a Lagrangian phase while sufficiently large, but as they break up, they are converted into a mixture with the diameter distribution of the droplet population modeled by S-Gamma. Last but not least, there is a Fluid Film on the out casing with impingement and stripping between the Lagrangian phase and the Multiphase Mixture. Schedule a meeting with CAEXPERTS and discover how to simulate complex multiphase phenomena - such as oil jet cooling in electric motors - with unprecedented precision and accessibility using Simcenter STAR-CCM+ . We're ready to help you turn engineering challenges into innovative solutions. WhatsApp: +55 (48) 98814-4798 E-mail: contato@caexperts.com.br