Gotion Uses InfluxDB to Manage Battery Data for Electric Vehicles

By Susannah Brodnitz / Product, Use Cases
Jul 27, 2022

As the world shifts to become greener and more sustainable, battery technology is of the utmost importance. Safe and efficient batteries drive environmentally friendly electric vehicles. Battery Management Systems (BMS) monitor batteries to give drivers information on battery status and mileage, and can also detect safety problems and send alerts.

Gotion is a company with offices around the world that specializes in battery technology. Its office in California works on BMSs. Engineers there work with battery data in order to design better and safer batteries. Some of the most important metrics they monitor include temperature, voltage, and current.

The challenge of battery data

Gotion’s batteries are sold in packs which are made up of hundreds of cells, and each pack has a BMS. Most BMSs capture voltage from those hundreds of cells, as well temperature and current from various sensors inside the pack. Each of these metrics has a different time resolution. A typical BMS collects voltage every 50 milliseconds, current every 10 milliseconds, and temperature every second.

Working with battery data has a unique set of challenges. It’s transmitted by wireless connections within cars which can go out for any number of reasons, including connectivity problems or when the vehicle is turned off. To deal with gaps in data, Gotion creates complex algorithms. The company needs a database that can handle large volumes of data and lets them apply their algorithms to it easily.

How InfluxDB helps

The Gotion team chose to store their time series data in InfluxDB for a few reasons. One is the availability of client APIs. They write algorithms in Python and MATLAB, and the ability to also work with InfluxDB in those same languages makes their system cohesive and simple.

The team stores data in InfluxDB and uses Grafana to create interactive dashboards and view data. Engineers can filter data by type and time window to choose what to apply the algorithm to. Gotion uses Airflow to send data from InfluxDB to Grafana, and to send back the results of the algorithm to InfluxDB.

Algorithm integration

One danger for batteries in electric vehicles is out of control overheating which can lead to fires. Because of the difficulties of working with this battery data, designing alerts around safe temperature thresholds doesn’t work well. Instad, the Gotion team developed an algorithm which sends alerts when the shape of temperature data starts to change.

Gotion’s algorithm has been successful at detecting temperature anomalies before thermal runaway occurs. With one set of temperature metrics, the algorithm was able to detect an anomaly based on data shape an hour and a half before an anomaly would have been detected based on passing the threshold. The detailed time series data Gotion collects and stores in InfluxDB allows it to make the batteries that power electric vehicles safer and more efficient.

To learn more about this use case, read the full case study.