Battery technology Porsche Engineering develops a digital twin of the high-voltage battery

Related Vendors

Porsche Engineering Group GmbH

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

SIMTOS (Seoul International Machine Tool Show)

The battery is considered the most important component of an electric vehicle. Not only in terms of capacity and thus range, but also because it significantly influences the residual value of an electric vehicle. In order to provide information on how battery cells and systems age, and the influence of user behavior on their lifespan, Porsche Engineering is working on a digital twin.

In order to optimize the high-voltage battery of electric vehicles, Porsche Engineering is pursuing the approach of the digital twin. "We need to understand how the cells behave in the long term in the field—without being able to draw on years of experience as with the combustion engine," explains Dr. Joachim Schaper, Head of AI and Big Data at Porsche Engineering. This is where the digital twin comes into play: The digital representation of the battery behaves exactly like the original and provides information about the expected aging process. It can also be used to improve the battery's lifespan and performance. AI experts from Porsche Engineering in Germany and the Czech Republic are therefore working hard on the digital battery twin.

Combination of models and field data

To create a digital twin of the battery, a modular, scalable framework for integrating existing and future model components is envisaged. The basis is a performance module that simplistically describes the electrical behavior of the battery and can build on approaches such as the resistor-capacitor model. In addition, there is a more complex electrochemical model that simulates the interaction between the anode, cathode, and electrolyte. The thermal model, on the other hand, allows one to foresee how the battery reacts to cold or heat.

The models are primarily based on laboratory tests with individual cells or cell modules and can therefore only predict to a limited extent how the battery behaves in the vehicle. That's why Porsche Engineering uses additional real field data from test vehicles or test benches where cells are measured. This information is potentially supplemented with data from the fleet.

Training AI Algorithms

Using field data, AI algorithms are trained to recognize patterns in customer usage behavior. For example, temperature or voltage deviations in individual cells can be signs of early wear and tear and anomalies. However, artificial intelligence can only recognize what there is a data basis for in the field. It cannot make statements about long-term aging effects because hardly any electric cars on the road are older than four years. Therefore, the company combines methods of artificial intelligence with existing model-based components.

First Function

A first function has emerged from the work on the digital battery twin: The Repair Prediction is based on a machine learning algorithm that monitors the battery data and warns of signs of wear or anomalies.

Work on the Digital Battery Twin began last year. Porsche Engineering has already created prototypes of the electrochemical and thermal models, which are now being combined with AI analyses. The challenge: data from vehicles with different thermal and charging systems must be combined, and the lab models are usually complex and require substantial computing power. The simulation models are gradually parameterized with field data to make them even closer to reality. Prototypical applications are expected later this year.

Looking into the future

The long-term goal is to not only create a general Digital Battery Twin, but also a digital representation of individual vehicle batteries. It could run in the cloud and, if desired, provide tips on how the user can extend the battery life with his behavior without suffering driving performance. Some factors known to positively affect durability are: the state of charge (SoC) should be kept constant between 30 and 70 percent and extreme outdoor temperatures should be avoided.

It is even conceivable to personalize the vehicle in the future using the digital twin. One example mentioned by Porsche Engineering is the analysis of the driving style and the adjustment of the BMS parameters to it, in order to reduce wear. Furthermore, digital twins could provide important insights for the development of new batteries in the future - beyond the automotive industry. (se)