Optimized fuel cell array energy management using multi-agent systems

Abstract

With the recent commercialization of modular fuel cell power plants, the interest for stationary fuel cells has grown over the last years. The main advantages of these energy sources are their limited environmental impact compared to traditional power sources such as coal- and gas-fired power plants, and at the same time, their ability to be controlled, as opposed to intermittent energy sources. However, controlling an entire array of fuel cells causes problems similar to the ones encountered by thermal power plant operators: how can the operation of such units be optimized? This article describes a method to help reach a better economic performance for fuel cell arrays, in which the individual performance of each fuel cell is considered. The method is based on a hybrid system consisting in a multi-agent system, an expert system and an optimization algorithm for optimally dispatching the load. By enabling the use of storage and the ability to automatically start and stop fuel cells, simulation results show that the energy management system can dramatically reduce running costs, as well as enable a very flexible control.