Hybrid Model to Calculate the State of Charge of a Battery

  1. María Teresa García Ordás 1
  2. David Yeregui Marcos del Blanco 1
  3. José Aveleira-Mata 1
  4. Francisco Zayas-Gato 2
  5. Esteban Jove 2
  6. José-Luis Casteleiro-Roca 2
  7. Héctor Quintián 2
  8. José Luis Calvo-Rolle 2
  9. Héctor Alaiz-Moretón 1
  1. 1 Universidad de León

    Universidad de León

    León, España

    ROR https://ror.org/02tzt0b78

  2. 2 Universidade da Coruña

    Universidade da Coruña

    La Coruña, España

    ROR https://ror.org/01qckj285

Hybrid Artificial Intelligent Systems: 16th International Conference, HAIS 2021. Bilbao, Spain. September 22–24, 2021. Proceedings
  1. Hugo Sanjurjo González (coord.)
  2. Iker Pastor López (coord.)
  3. Pablo García Bringas (coord.)
  4. Héctor Quintián (coord.)
  5. Emilio Corchado (coord.)

Publisher: Springer International Publishing AG

ISBN: 978-3-030-86271-8 978-3-030-86270-1

Year of publication: 2021

Pages: 379-390

Congress: Hybrid Artificial Intelligent Systems (HAIS) (16. 2021. Bilbao)

Type: Conference paper


Batteries are one of the most important component in an energy storage system; they are used mainly in electric mobility, consumer electronic and some other devices. Nowadays the common battery type is the liquid electrolyte solution, but it is expected that in some year, the solid state batteries increase the energy density. Despite the type of batteries, it is very important that the user knows the energy that remains inside the battery. The most used ways to calculate the capacity, or State Of Charge (SOC), is the percentage representation that takes into account energy that can be stored, and the remained energy. This research is based on a Lithium Iron Phosphate (LiFePO4) power cell, because it is commonly used in several applications. This paper develops a hybrid model that calculate the SOC taking into account the voltage of the battery and the current to, or from, it. Moreover, there has been checked two different clustering algorithms to achieve the best accurate of the model, that finally has a Mean Absolute Error of 0.225.