Institutional Repository [SANDBOX]
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Dynamic modelling and simulation of a high-temperature PEM fuel cell

Boula Foteini

Full record


URI: http://purl.tuc.gr/dl/dias/EA76FDE5-CAA5-485F-A3F6-D70C813EACC0
Year 2024
Type of Item Diploma Work
License
Details
Bibliographic Citation Foteini Boula, "Dynamic modelling and simulation of a high-temperature PEM fuel cell", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2024 https://doi.org/10.26233/heallink.tuc.100504
Appears in Collections

Summary

Over the last few years, hydrogen fuel cells have emerged as a promising alternative for clean and efficient energy conversion. Fuel cells convert the chemical energy of a fuel like hydrogen into electricity through a chemical reaction with an oxidizing agent, and only water and heat as byproducts. The advantages of this newtechnology compared to conventional energy sources are significant, with some of them being higher efficiency, lower emissions, and flexibility in fuel selection.There are various types of fuel cells, such as Proton Exchange Membrane Fuel Cells (PEMFC), Alkaline Electrolyte Fuel Cells (AFC), and others. Proton Exchange Membrane Fuel Cells (PEMFC) have the widest range of applications, making them the most commonly used. This type of fuel cell was first developed byGeneral Electric in the 1960s for NASA’s first manned space vehicles.Hydrogen fuel cells can be characterised as a promising technology compared to traditional combustion engines due to their higher efficiency, lower emissions, and flexibility in fuel selection. However, challenges such as cost, durability, and reliability need to be taken into consideration to enable the establishment of these fuelcells.The main principle that connects a fuel cell’s functionality with the output voltage is the V-I curve. In ideal conditions, the output voltage aligns with the open circuit voltage, Vo. However, real-world conditions involve activation losses, resistive losses, and limitations of mass transfer. This equation was not studied in detail inthe present thesis. Additionally, the operating voltage in such conditions will contain losses, determined by a parametric equation based on the operating temperature and oxygen concentration.The focus of this study was the dynamic behaviour of a high temperature fuel cell by including mass and energy balance differential equations (1st order). The circuit’s mathematical description includes ohmic losses and concentration losses. Ohmic losses are due to resistance during ion and electron transfer. Concentrationlosses are related to mass transfer and occur when the fuel cell operates at a higher current.To simulate the system, a dynamic model was developed in MATLAB environment, that consisted of a non-linear model of the above-mentioned 13 first-order differential equations and various parameters. From that, with the use of the Taylor factorization, the model was linearised and then the state-space model wasproduced. The state space model is a contemporary control system design methodology that overcomes the limitations of function-based methods and provides insight into the system’s internal state while taking into account the system’s initial conditions.The fuel cell system was simulated using three scenarios: non-linear code, linear code, and state-space model.The goal was to compare results and assess the alignment of the linear and the state-space models with the non-linear model while considering constants for consistency. The analysis was illustrated through graphs and table comparison.To ensure the system follows the changes in fuel and coolant flow, various cenarios were simulated for the three models. These scenarios were: increase and decrease by 30% in hydrogen and coolant input flows, increase in hydrogen input flow by 40% with a simultaneous decrease in coolant input flow by 20%, decrease in hydrogen input flow by 40% with a simultaneous increase in coolant input flow by 20%, increase and decrease by 30% in hydrogen input flow, and increase and decrease by 30% in both hydrogen and coolant input flows. The latter two scenarios were only executed for the state-space model.After executing the simulations for the base case scenario and all other variations, the results were displayed in graphs. The outcomes were those expected with no serious discrepancies among the models. It can be concluded that the system responds accordingly to various changes in both hydrogen and coolant flow input.To further develop this study, various controllers can be introduced to control the system in a more automated way.

Available Files

Services

Statistics