The student Markel Azkue Etxeandia obtained an EXCELLENT grade with International Doctorate mention

Back

The student Markel Azkue Etxeandia obtained an EXCELLENT grade with International Doctorate mention

THESIS

The student Markel Azkue Etxeandia obtained an EXCELLENT grade with International Doctorate mention

2023·12·18

$titulo.getData()

  • Thesis title: Development and validation of Li-lon battery state algorithms capable of adapting to new chemistries

Court:

  • Presidency: Hendrik Johannes Bergveld (Eindhoven University of Technology)
  • Vocal: Elisa Braco Sola (BeePlanet Factory)
  • Vocal: David Anseán González (Universidad de Oviedo)
  • Vocal: Vincent Heiries (CEA-LETI)
  • Secretary: Laura Oca Perez (Mondragon Unibertsitatea)

Abstract:

Lithium-ion batteries are crucial in modern applications, ranging from devices like phones and laptops to electric vehicles. The increase in demand is driven by the shift toward renewable energy and electric mobility. The research focuses on estimating the state of charge (SoC) and state of health (SoH) of these batteries.

The study aims to develop flexible SoC and SoH estimators using neural networks and transfer learning. An examination of existing algorithms is conducted, and a five-stage methodology is proposed for model creation.

The results demonstrate that transfer learning models consistently outperform those built from scratch, with significantly lower errors and reduced data requirements. This combination of neural networks and transfer learning is effective for estimating SoC and SoH in data-constrained scenarios, delivering high-quality results.