Date of Graduation


Document Type



The objective of this research is to analyze and control a solid oxide fuel cell (SOFC) used as a distributed generator. First, a detailed dynamic model of a SOFC is developed, which can be used for small signal and transient stability studies. The model based on electrochemical and thermal equations accounts for temperature dynamics and output voltage losses. The output voltage response of a standalone fuel cell plant to different and fast load variations is simulated to illustrate the dynamic response of SOFC to fast and slow perturbations. The proposed model is incorporated with a MATLAB based power system stability package named the Power Analysis Toolbox (PAT). Second, the developed SOFC model is used to analyze the stability of distribution systems containing fuel cells as distributed generators (DGs). The SOFC power/utilization control methods are explained and a controller is designed to avoid cell damages. The impact of the developed SOFC model on the stability of a 13-bus distribution system, when it is islanded from the substation, has been tested. Controls have been designed to improve reliability and power delivery of distribution systems. Frequency fluctuations after islanding are considerably reduced compared to the case when only gas turbines are used as DGs. It is demonstrated that DGs, namely a SOFC and a microturbine (MT), could cause instability of electric power distribution systems following fault conditions. A control algorithm is proposed to enhance transient stability in case of large disturbances. The control algorithm consists of (1) disconnecting the SOFC during the fault, and (2) implementing fixed structure decentralized SOFC reconnection controllers and power system stabilizer for the MT generator. Finally, the concept of a distribution system that has enough generation to track its load without the help of a substation is presented. Two control loops are proposed for the SOFCs to guarantee that SOFC is protected and to track load changes and regulate the frequency. These two respective loops are called primary and secondary loops. Tuning of the secondary loop controller parameters are performed using genetic algorithms. A Distribution System Error (DSE) is introduced to formulate the frequency control problem.