Modeling and Simulation for Meeting the Challenges of Battery Design

The biggest challenges for battery design are energy density, power density, charging time, life, cost, and sustainability. Modeling and simulation are very efficient methods that can assist researchers, developers, and designers in meeting these challenges.

4 min read
The complexity of today’s batteries demands flexible computational tools.

The following is an excerpt from Multiphysics Simulation 2017.

By Ed Fontes

Energy density is limited by the battery’s chemistry, which even without losses limits the theoretical energy density. The chemistry is defined by the electrode material and the composition of the electrolyte. Lithium-air batteries get close to the energy density of gasoline, which is probably close to the maximum energy density for a battery. However, the components required for thermal management and current collection contribute to the total weight of the battery system. The design of these components can substantially influence the energy density of a battery system.

Figure 1. Multiphysics simulation of the temperature profile in a liquid-cooled battery pack. Figure 1. Multiphysics simulation of the temperature profile in a liquid-cooled battery pack. Fluid flow and temperature are modeled in 3D and a lumped 1D model of the Li-ion battery is used to calculate the heat source.


The power density of a battery is important for the efficiency of electric vehicles. A high power density is required to recapture high amounts of energy in a short time during

regenerative braking or fast recharge. This gives a difficult optimization problem, since the system has to cope with very high current densities during recharge and relatively low current densities during discharge. It also relates to the design of the thermal management and the current collectors mentioned earlier. In addition, the design of fundamental battery components such as the electrodes, separator, and electrolyte are of great importance for power density.


Life is a major consideration where safety and reliability are closely related. Discharge, wear, and failure should occur slowly and in a controlled and transparent way. This is not only an issue of the chemistry of the battery, but also of the design, since uneven current density distribution and poor control of discharge/recharge and of the thermal management system may accelerate wear and increase the risks of failure. Short-circuits formed by metal deposition may be responsible for decrease in performance as well as an increased risk for runaway heating. Technologies for state-of-health monitoring are required in order to continuously assess the state of the battery system and the risks of failure.


The manufacturing process for high-power batteries and electric powertrains is not as optimized as for mechanical powertrains for combustion engines. There is a larger potential in productivity gains and decreased costs by large-scale production in the manufacturing process for the battery components.


The development of new batteries has to include the aspect of sustainability. There has to be a strategy for mining, recycling, producing, and disposing of new battery types. This is primarily a legal matter for governments, but also a commercial consideration for battery manufacturers and automotive companies.


The understanding and optimization of fundamental components of the battery; such as electrodes, electrolyte, and separator; can be accelerated using modeling and simulations. The systems for thermal management, current collection, and state-of-health monitoring can also be developed with high-fidelity multiphysics simulations.

Figure 1 shows a model for the optimization of the channels in the cooling plates of a battery pack. The is a common application in the automotive industry. For example, Fiat Research Center uses mathematical modeling for studying thermal management of pouch cells for hybrid vehicles1.

Figure 2. The goal with this application is to explain experimental electrochemical impedance spectroscopy (EIS) measurements and to show how Figure 2. The goal with this application is to explain experimental electrochemical impedance spectroscopy (EIS) measurements and to show how you can use the model and the measurements to estimate the properties of lithium-ion batteries. The application takes experimental data from EIS measurements as input, simulates these measurements, and then runs a parameter estimation based on the experimental data.

Fundamental studies of battery components as well as the development of state of health methods can be very efficiently carried out by combining experimental measurements of electrochemical impedance spectroscopy (EIS) with mathematical models, see the article from the French research institute CEA2. Figure 2 shows an app, where experimental data can be imported and used in a physics-based model of EIS. This allows for the estimation of parameters such as activity of the electrodes, surface area, electrical conductivities of the different components, mass transport properties of reactants and products, and state of charge of the electrodes.

Click here to read the 2017 edition of Multiphysics Simulation and learn how mathematical modeling and multiphysics simulation are being leveraged as a powerful tool in many other industries.


1. The Thermal Management of Li-ion Battery Packs. Michele Gosso, Antonio Fiumara. Fiat Research Center, Orbassano, Italy. COMSOL News (2012), 48-49.

2. Highly Accurate Li-ion Battery Simulation. Mikael Cugnet. French Atomic and Alternative Energy Commission (CEA), France. COMSOL News (2013), 44-45.

The Conversation (0)

The Future of Deep Learning Is Photonic

Computing with light could slash the energy needs of neural networks

10 min read

This computer rendering depicts the pattern on a photonic chip that the author and his colleagues have devised for performing neural-network calculations using light.

Alexander Sludds

Think of the many tasks to which computers are being applied that in the not-so-distant past required human intuition. Computers routinely identify objects in images, transcribe speech, translate between languages, diagnose medical conditions, play complex games, and drive cars.

The technique that has empowered these stunning developments is called deep learning, a term that refers to mathematical models known as artificial neural networks. Deep learning is a subfield of machine learning, a branch of computer science based on fitting complex models to data.

Keep Reading ↓ Show less