Numerical investigation of micro solid oxide fuel cell performance in combination with artificial intelligence approach

Parastoo Taleghani, Majid Ghassemi, Mahmoud Chizari

Research output: Contribution to journalArticlepeer-review

Abstract

The current study presents a multiphysics numerical model for a micro-planar proton-conducting solid oxide fuel cell (H-SOFC). The numerical model considered an anode-supported H-SOFC with direct internal reforming (DIR) of methane. The model solves coupled nonlinear equations, including continuity, momentum, mass transfer, chemical and electrochemical reactions, and energy equations. Furthermore, The numerical model results are used in artificial intelligence (AI) models, the K-nearest neighbour (KNN) and, artificial neural network (ANN), to predict the current density and power density of the H-SOFC. The results show that increasing the air-to-fuel (A/F) ratio decreases the current density and overall cell power. In particular, improvements in power and current density observed in H-SOFC when the A/F ratio is set to 0.5, resulting in a respective increase of 2 % and 7 % compared to the initial state at A/F = 1. With an error rate of less than 1 % and an R-score of around 99 %, the ANN model shows good agreement with the numerical results.

Original languageEnglish
Article numbere40996
Pages (from-to)1-17
Number of pages17
JournalHeliyon
Volume10
Issue number24
Early online date6 Dec 2024
DOIs
Publication statusPublished - 30 Dec 2024

Keywords

  • Artificial intelligence
  • Artificial neural network
  • Micro solid oxide fuel cell
  • Numerical model
  • Proton-conducting electrolyte

Fingerprint

Dive into the research topics of 'Numerical investigation of micro solid oxide fuel cell performance in combination with artificial intelligence approach'. Together they form a unique fingerprint.

Cite this