Electric vehicles (EVs) offer the possibility to address the world’s transportation demands in a more environmentally sustainable fashion. Mass adoption can help to reduce reliance on fossil fuels but the lithium-ion (Li-ion) chemistries that represent current state-of-the-art battery technology still present unique challenges to designers and engineers — a chief development priority being to limit the potential for battery fire.

To achieve vehicle manufacturers’ ambitious EV adoption goals, it is necessary to bolster and improve the safety of Li-ion batteries by better understanding all of the complex, interconnected aspects of their behavior across both normal and extreme duty cycles.

Altair is focused on developing a comprehensive understanding of automotive battery safety issues that it has named the Altair Battery Designer project. It combines innovative design methods and tools to model and predict mechanical damage phenomena as well as thermal and electro-chemical runaway. Altair has developed an efficient way to calculate mechanical and short-term thermal response to mechanical abuses, providing accurate computational models and engineer-friendly methods to design a better battery.

This was made possible through the following collaborations:

  • For the mechanical aspect, with the MIT Battery Modeling Consortium, which involves several automotive OEMs, battery manufacturers, and US governmental research organizations.

  • For the mechanical and thermal runaway coupling aspect, with the Vehicle Energy & Safety Laboratory (VESL) and North Carolina Motorsports and Automotive Research Center out of the University of North Carolina, Charlotte (UNCC), under the leadership of Professor Jun Xu.

These collaborations provided both critical industry expertise as well as a wide range of reliable experimental data to validate Altair’s methods.

Challenges

Given the complexity of an individual battery cell’s structure and moreover, of a module or a full battery pack, homogenization techniques are required to keep the model size and CPU time compatible with the aggressive time schedules of industrial design cycles.

Based on the mechanical and electro-thermal experimental tests of battery cell components (anode, cathode, separator, etc.), homogenized properties of the cell were identified using Altair Radioss™ and Altair Flux™ software.

The mechanical behavior of the homogenized element is modeled as an orthotropic elasto-plastic material. In addition, a failure model is used with orthotropy, strain rate, state of charge, and temperature effects. Post-rupture properties are also considered.

Figure 1. Simulation vs test results of a battery cell hemispherical indentation.

Thanks to the contribution of Dr. Elham Sahraei from Temple University College of Engineering and co-director of the MIT Battery Modeling Consortium (2016-2019), an accurate method was established for identifying the mechanical properties of the battery cells, including its possible failure, using advanced numerical modeling techniques.

Figure 1 shows the correlation between the simulation’s prediction and actual field data from a prismatic cell indentation test.

Problem Formulation

Used to predict whether the vehicle adheres to regulatory crash and impact regulations, Altair’s homogenization method enables the accurate simulation of a full electric vehicle, including a realistic deformable battery pack, to be run in less than one night. Once implemented in an optimization process, the design of the entire vehicle can be optimized, leading to a high degree of safety at a production low cost. The deformations at the battery pack for a pole impact EuroNcap test are calculated in the EV model in Figure 2.

Figure 2. Battery mechanical response to a vehicle pole impact.

The next step consists of calculating the thermal effect caused by deformations of materials inside the battery components to determine the risk of thermal runaway. A method has been designed and validated in collaboration with Professor Jun Xu’s team at UNCC using a cylindrical cell.

Figure 3. Power loss vs strain curve.

Altair Flux is used to perform the electro-thermal characterization of the homogenized element. Internal short circuits are considered, based on a compressive strain criterion, and the power loss in the material is extracted from the tests. Figure 3 shows the power loss variations as a function of the strain in the material. The ensuing change of resistivity is established based on experimental data as shown in Figure 4.

Once the homogenized element is fully characterized, an explicit transient nonlinear dynamic calculation is performed using Radioss to obtain the mechanical response of the cell. A transient heat transfer analysis is accomplished using Altair Optistruct™, exploiting the deformed shape of the battery calculated by Radioss and the power losses calculated from the homogenized element strain values.

Figure 4. Experimental ISC measurement.

Altair Battery Designer Process Description

Experiments and Results

The process involves several complex operations: numerical multiphysics simulations including highly nonlinear structural dynamic analysis, transient thermal analysis, and electro-thermal interactions. It is therefore fundamental to validate the accuracy of the results and the predictivity of the process on simplistic applications before conducting it on a full-scale battery or electric vehicle design.

Figure 5. Test setup and structural simulation models.

The validation was performed on an NCR 18650B battery cell, 30% SOC. Two tests were performed: compression and indentation (Figure 5). Both voltage/force and temperature/impactor displacement variables were measured. The calculation times were completed in just a few minutes on a laptop for the mechanical analysis and few seconds for the thermal analysis.

The different force, voltage, and temperature curves histories from the tests are plotted on Figure 6. The mechanical rupture of the cell corresponds to the short circuit event with the voltage drop and the beginning of the temperature increase.

Figure 6. Experimental results.
Figure 7. Numerical results vs experimental curve.

The mechanical and thermal simulation results (Figure 7) all correlate well with the physical tests, which demonstrates that the process properly accounts for the multi-physics behavior. Temperature increases within the battery pack can also be calculated at the moment of a vehicle crash. Using Altair’s multiphysics methodology, the electric vehicle model was submitted to a lateral pole impact test. The instantaneous temperature changes due to the mechanical deformations of the cells inside the battery pack were considered.

The melting temperature of the separator in the cells is 260 °C. Any areas showing dramatic temperature increases at the point of impact tell analysts where design modifications may be needed. Figure 8 highlights two locations into the battery reaching 160 °C when the state of charge is at 30 percent. Increasing the state of charge could increase the risk of thermal runaway in these two zones, so it would be recommended to perform further structural optimization studies to avoid compression of cells in these areas.

Figure 8. Temperature increases at the end of the pole impact test.

This method allows designers to quickly identify potential safety issues early in the battery design process, shortening development time and ultimately leading to the design of safer, more cost-effective, and better-performing electric vehicles.

Conclusion

With the increased use of Li-ion batteries in the transportation world, verifying battery safety — even in abuse cases such as impact and shock — has become an important issue in our day-to-day life.

Using Altair Battery Designer’s tools and simulation methods, it is now possible to accurately predict the short-term mechanical and thermal behaviors of a battery under abusive loading in a short period of time. This unique process can be applied to multiple load cases such as crash, crush, deceleration, and impacts with and without penetrations.

This article was written by Jean-Baptiste Mouillet – Multiphysics Solutions Director; Marian Bulla – Program Manager, Material Data; Jean-Michel Terrier – VP, Radioss Worldwide Business Development; and Patrick Lombard – Lead Application Specialist Manager, Altair Engineering (Troy, MI). For more information, visit here .