Data-Driven Modelling for Health Estimation of High-Voltage Battery Systems

Master Thesis in Systems, Control and Mechatronics

An accurate determination of battery health and life prediction is essential to ensure reliable, efficient and durable battery performance along the full lifetime of a vehicle. This thesis builds on achievements with data-driven modelling to determine the behaviour of complex dynamical systems through machine learning techniques. The conducted survey over a range of model techniques, from standard baseline up to state-of-the-art approaches, indicates the power and flexibility of data-driven models in the directions of per vehicle and fleet use cases.

Technologies used: ClickhouseDB, Numpy, Pandas, Jupyter Lab, Keras, Tensorflow
Publication: Download the final publication here!
Presentation: corresponding presentation event here!

Recommended citation (APA 6th reference style):
Rauh, Lukas. (2020). Data-Driven Modelling for Health Estimation of High-Voltage Battery Systems (Master Thesis, Chalmers University of Technology, Gothenburg, Sweden) Retrieved from https://hdl.handle.net/20.500.12380/301548