Promoting excellence in mobility engineering

  1. FISITA Store
  2. Technical Papers

Fuel Cell Polarization Curve Neural Network Estimator for Automotive Applications
JUMV/EAEC05YU-EF05

Authors

M. Milacic - Sustainable Mobility Technologies, Ford Motor Company

Abstract

Keywords

Self organizing map, Fuel cells, Polarization curve

Abstract

Even before record prices of crude oil [1] and growing environmental awareness increased U.S. government commitment toward renewable fuels and sustainable technologies [2], the automotive industry successfully demonstrated fuel cell technology in transportation applications, ranging from small passenger cars to buses [3]. A fuel cell stack polarization curve (current-voltage relationship curve) is often used to evalu- ate fuel cell stack characteristics, such as maximum power available. Based on the polarization curve some internal problems can be diagnosed. The polarization curve is usually obtained by following a procedure to obtain voltage readings at various load points. In this paper a method of acquiring the stack polarization curve in order to monitor perfor- mance in a Ford Focus fuel cell hybrid vehicle is presented. This method is based on a self organizing map and uses information acquired during regular driving to estimate the polariza- tion curve. The advantage of this approach is that it does not require any special procedure or prede¯ned drive pro¯le.

Add to basket

Back to search results