A systems approach for sizing a stand-alone residential PEMFC power system

Selamoğulları U. S. , Willemain T., Torrey D. A.

JOURNAL OF POWER SOURCES, vol.171, pp.802-810, 2007 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 171
  • Publication Date: 2007
  • Doi Number: 10.1016/j.jpowsour.2007.06.024
  • Journal Indexes: Science Citation Index Expanded, Scopus
  • Page Numbers: pp.802-810
  • Keywords: fuel cell, residential load, stochastic, sizing, power systems, stand-alone, ENERGY-STORAGE SYSTEMS, FUEL-CELL, PERFORMANCE, BOOTSTRAP


Polymer electrolyte membrane fuel cells (PEMFCs) show great promise in portable, automotive, and stationary applications. They have reached the test and demonstration phase in automotive and power markets today. This paper is focused on a stand-alone residential PEMFC power system that provides the electricity needs of the house. A novel stochastic sizing methodology is developed that considers both fuel cell system dynamics and residential load dynamics in overall system sizing for the stand-alone residential fuel cell power system. Understanding the nature of demand side is critical in stand-alone system sizing. Thus, experimental measurements have been completed to capture the load side dynamics in detail. No such data is found in the current literature. The Threshold Bootstrap method is used to model the residential load demand and to produce many realistic load profiles. Matlab/Simulink is used to run system simulations to determine system sizes based on parameters defined through a designed experiment. Comparison between the proposed sizing method and a possible worst case scenario sizing is given. The new sizing methodology can be used together with sophisticated demand analysis programs to obtain customized sizing for each user as stand-alone power systems become more viable. (C) 2007 Elsevier B.V. All rights reserved.