Optimal design of proportional-integral controllers for stand-alone solid oxide fuel cell power plant using...

Area of Science:

• Energy Systems Engineering
• Control Systems Theory
• Computational Intelligence

Background:

• Stand-alone solid oxide fuel cell (SOFC) systems require robust control for stable operation.
• Small signal dynamic response is crucial for tracking load variations and maintaining system stability.
• Proportional-Integral (PI) controllers are commonly used but require optimal tuning for enhanced performance.

Purpose of the Study:

• To apply the differential evolution (DE) algorithm for optimal tuning of PI controllers in a stand-alone SOFC system.
• To improve the small signal dynamic response of the SOFC system to load current variations.
• To evaluate the effectiveness of DE compared to other optimization algorithms.

Main Methods:

• Derivation of the small signal model for the SOFC system.
• Design of two PI controllers for hydrogen and oxygen partial pressure feedback loops.
• Formulation of the controller design as an eigenvalue-based objective function minimization problem.
• Application of the Differential Evolution (DE) algorithm to find optimal PI controller gains.

Main Results:

• Eigenvalue and time-domain simulations confirmed the effectiveness of DE-tuned PI controllers.
• DE outperformed Particle Swarm Optimization (PSO) and Invasive Weed Optimization (IWO) in achieving optimal controller gains.
• Statistical analyses (Kolmogorov-Smirnov test, t-test) confirmed the superiority of DE.

Conclusions:

• The Differential Evolution (DE) algorithm is a highly effective tool for optimizing PI controllers in SOFC systems.
• DE provides a statistically significant advantage over PSO and IWO for this control problem.
• Optimal PI control enhances the small signal dynamic response and load-tracking capabilities of stand-alone SOFCs.