Improving battery performance with Entroview

, , ,

To optimize the production of batteries for electric vehicles, the start-up Entroview develops diagnostic solutions to speed up the testing phase. The young company’s primary targets are gigafactories and industries specialized in battery design.

With the growth of the electric vehicle market, battery production is a key issue for vehicle manufacturers and gigafactories. To guarantee the production of high-performance lithium batteries, Entroview, a start-up created in 2021 and which will be present at the VivaTech 2022 fair, proposes a diagnostic solution for the design phase. “The batteries we use every day are complex chemical systems,” says Sohaïb El Outmani, CTO of Entroview. “We must be able to detect inefficient batteries during production and understand the reasons why,” he explains.

The start-up, which was incubated at Télécom Paris Novation Center, develops software to reduce the testing time on batteries. It would allow tests, which currently take several days, to be reduced to a process of just three hours, saving time and cost for the manufacturer. This method would also make it possible to estimate the quality of the cells (the units that store the electricity in a battery) using current and temperature measurements. Based on this data, the algorithm developed by Entroview could measure entropy, or the organization of particles in a system. “The algorithm we are developing measures the entropy of the lithium atoms in a battery,” says Sohaïb El Outmani. “In a battery, it is preferable to have low entropy variation,” he explains. High entropy variation tends to be associated with the deterioration of the battery.

In a lithium battery, entropy is related to the chemical composition, the size of the powders incorporated in the cells and the state of deterioration. The program developed by Entroview is designed to determine the deterioration mechanisms at play in the batteries. Some mechanisms are known to be specific to the end of life of batteries, making it possible to select the best performing batteries using physical analyses carried out by the software. 

A solution adapted to each need

The software is based on “a mixture of machine learning algorithms and data acquisition, including physical data.” Using this data, the algorithm identifies the different types of batteries in the test phases according to their entropy profile. “There are different types of lithium batteries, which all have different chemical compositions” says the entrepreneur. With a diagnostic specific to each type, the manufacturer can make the right decisions to improve their production.

Entroview will also offer consulting services to help companies understand the diagnostics. However, as the software carries out the different analyses, it will incorporate more and more data, enabling it to recognize ever more varied problems with increasing accuracy. This will minimize the need for expert intervention to understand the diagnostics.

The start-up also plans to develop an application to track the condition of batteries during their use. “The aim is to integrate the software into vehicles to diagnose the health and state of charge of batteries.” For example, the software could predict the remaining life of a battery based on the user’s current usage pattern. The solution could also suggest journeys to optimize battery life, “which could be of interest to transport companies that manage bus fleets,” says Sohaïb El Outmani.

According to the CTO, “if we were already able to predict battery life, it could be extended by 50%.” The solutions developed by the start-up are currently mainly intended for gigafactories, and it will be a few more years before the software is commercialized. “European gigafactories will be up and running in Europe within three years. Our aim is for our solution to be operational by then,” says Sohaïb El Outmani.

Rémy Fauvel

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *