CUSTOMER STORY | AUTOMOTIVE
Volvo isolates the variables that degrade EV batteries from more than 1,400 sensors
To design an electric vehicle battery that is both efficient and durable, Volvo pushes prototypes through rapid thermal shocks. TIMi processes the telemetry of more than 1,400 simultaneous sensors to isolate what causes degradation.
OVERVIEW
The challenge.
Make sense of a massive flow of telemetry, and find the hidden variables.
Extreme thermal shocks
Prototypes are pushed to their limits in the lab, with rapid thermal shocks that simulate accelerated aging.
More than 1,400 sensors
Volvo needed to isolate which hidden variables cause battery degradation.
SOLUTION
What was done.
Raw stress-test data turned into engineering decisions.
R&D analytics
Predictive modeling processes the massive sensor data to isolate the variables that affect battery performance under thermal stress.
Iterative design loop
The insights dictate engineering and material choices for the next generation of batteries, in a continuous improvement cycle.
Factory floor
Proven in the R&D lab, the predictive capability was extended to the manufacturing floor, for heavy industrial tools.
RESULTS
What changed.
Degradation drivers identified, and a method that moved from the lab to the factory.
Degradation variables
The variables responsible for battery degradation are isolated.
Next generation
The results guide the engineering and material choices of the next batteries.
Beyond the lab
The same approach now supports the performance and maintenance of heavy industrial tools.
Testimonials
Their feedback
“Very fast on big data sets compared to other software, I have run 100’s of models for each of my difficult to solve root cause analysis projects, where big money was at stake in warranty costs.”
Artur G. – Data Scientist, Volvo Cars
USE CASES BEHIND THIS STORY
Read the method.
Asset Degradation Modeling
Model how the condition of an asset evolves with use, and estimate its remaining life.
Sensor Data Modeling at Scale
Model thousands of sensor variables together on a standard server.
IoT Data Processing
Receive and process MQTT sensor streams on one or two standard servers.
SEE FOR YOURSELF
Test it on your own data.
Run TIMi on your own data with a free trial, or talk to us about your case.
