A proposed artificial neural network model for PEM fuel cells

Sari, Ali; Balıkçı, Abdulkadir; Taşkın, Sezai; Aydın, Serkan

doi:10.1109/eleco.2013.6713832

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following steps should be taken for the implementation of Neural Network [30] 6) where 'd' is the parameter of slope and 'u' is the value from the input layer. This non-linear transfer function in the hidden layer helps in the dynamic modeling of fuel cells and gives linear output.…”

Section: Designing Ann-based Controllermentioning

confidence: 99%

Flow Control of Hydrogen Fuel in Pem Fuel Cell Using Soft Computing Techniques

QAISER¹,

ASGHAR²,

JAFFERY³

et al. 2021

JOR

View full text Add to dashboard Cite

Fuel cells transform the chemical energy of hydrogen directly into electrical energy without ignition or thermal processes. Their behavior is defined based on electrochemistry and thermodynamics that involves complex computations in their mathematical model. This problem of modeling can be resolved by using soft computing techniques. Fuel cells are effective, versatile and silent devices that can provide power to many applications¸ from portable electronic devices to automobiles, to electrical grids across the nation. Due to the nonlinear process of a fuel cell, fuzzy logic, neural network, and Neurofuzzy controllers are suitable for regulating input gasses flow rate to get appropriate electrical power according to load demand. This paper describes aMATLAB / Simulink model of 1KW, 28.8V DC power PEM fuel cell for controlling hydrogen flow rate to the fuel cell stack using fuzzy logic, neural network, and Neuro-fuzzy controllers. The output performance of controllers is compared based on their efficiency and utilization. Simulation results showed that the Neuro-Fuzzy controller provides good performance for the purging process of hydrogen

show abstract

Section: Designing Ann-based Controllermentioning

confidence: 99%

Flow Control of Hydrogen Fuel in Pem Fuel Cell Using Soft Computing Techniques

QAISER¹,

ASGHAR²,

JAFFERY³

et al. 2021

JOR

View full text Add to dashboard Cite

show abstract

Improved performance of PEM fuel cells stack feeding switched reluctance motor using multi-objective dragonfly optimizer

El-Hay

El‐Hameed

El‐Fergany

2018

Neural Comput & Applic

View full text Add to dashboard Cite

A computationally efficient and high-fidelity 1D steady-state performance model for PEM fuel cells

Zhao

Li²,

Shum³

et al. 2023

J. Phys. Energy

View full text Add to dashboard Cite

The performance of a proton exchange membrane (PEM) fuel cell is determined by many factors, including operating conditions, component specifications, and system design, making it challenging to predict its performance over a wide range of operating conditions. Existing fuel cell models can be complex and computationally demanding or may be over-simplified by neglecting many transport phenomena. Therefore, a high-fidelity and computationally efficient model is urgently needed for the model-based control of fuel cells. In this study, semi-implicit multi-physics numerical models have been established, taking the mass, momentum, reactants, liquid water, membrane water, electrons, ions, and energy in all fuel cell components into account. The developed 1D model is of high fidelity by incorporating the two-phase flow, non-isothermal effect, and convection, and is still computationally efficient. These models are validated against data from an auto manufacturer with good agreements, and the computing efficiency is evaluated on a modest laptop computer. The modeling results suggest that the two-phase flow model exhibits better prediction accuracy than the single-phase flow model when reactants are fully humidified, while under low humidity conditions, the two models present equivalent performance as liquid water does not exist in the fuel cell components. The results also suggest that the maximum convective/diffusive ratio of H2, O2, and vapor mass fluxes can be 12%, 5.3%, and 35%, respectively, which are ignored in most diffusion-dominant models. The developed models are computationally efficient, requiring only 0.56 s and 0.26 s to simulate a steady-state operation of fuel cells for the two- and single-phase flow models, respectively. This implies that the developed models are suitable for the control of PEM fuel cells.

show abstract

A proposed artificial neural network model for PEM fuel cells

Cited by 3 publications

References 14 publications

Flow Control of Hydrogen Fuel in Pem Fuel Cell Using Soft Computing Techniques

Flow Control of Hydrogen Fuel in Pem Fuel Cell Using Soft Computing Techniques

Improved performance of PEM fuel cells stack feeding switched reluctance motor using multi-objective dragonfly optimizer

A computationally efficient and high-fidelity 1D steady-state performance model for PEM fuel cells

Contact Info

Product

Resources

About