Search

Browse

• Communities

My Space

• Login

Modeling and control of a PEM fuel cell for use in hydrogen vehicles

Charalampous Nikolaos

Full record

---

Summary

This thesis focuses on the dynamic modeling and control of a Proton Exchange Membrane (PEM) fuel cell stack intended for hydrogen-powered vehicle applications. PEM fuel cells are increasingly important in clean energy technologies due to their high efficiency, low emissions, and rapid dynamic response. The accurate modeling and design of appropriate control strategies are thus critical to optimizing their performance and integration into energy systems.The system is first modeled using a set of nonlinear differential equations, which describe the core state variables of the fuel cell stack, including chemical concentrations, thermal dynamics, current production, and reactant flows. This nonlinear model is then linearized using a Taylor series expansion around steady-state points, resulting in a simplified linear form. Subsequently, the model is expressed in state-space representation, enabling the application of advanced control methodologies.Following model development, simulation results are presented for each of the three models—nonlinear, linear, and state-space. These models are compared under nominal and perturbed conditions by introducing ±30% changes in hydrogen input flow and coolant mass flow rate (air). The consistency of the results confirms the equivalence and accuracy of the state-space model, making it suitable for control design.In the final phase, two closed-loop control problems are addressed: regulating the output current through hydrogen input and managing system temperature via coolant flow. Controllers of type P, PI, and PID are designed using the Ziegler–Nichols and Tyreus–Luyben tuning methods. Their performance is assessed using standard performance criteria (ISE, IAE, ITSE, IATE). Simulation results indicate that the PID controller tuned via Tyreus–Luyben performs best for current control, while the Ziegler–Nichols PID configuration is more effective for temperature regulation.

Available Files

Services