A management of multi-stacks fuel cell systems is proposed to extend systems useful life in a Prognostics and Health Management (PHM) framework. The problem consists in selecting at each time which fuel cell stacks have to run and which output power has to be chosen for each of them to satisfy a load demand as long as possible. Multi-stacks fuel cell system useful life depends not only on each stack useful life, but also on both the schedule and the operating conditions settings that define the contribution of each stack over time. As the impact of variable operating conditions on fuel cell lifetime is not well-known, a simplified representation of fuel cell behavior under wear and tear is used to estimate the available outputs over time and their associated Remaining Useful Lives (RUL). This health state prognostics model is configured to suit to Proton-Exchange Membrane Fuel Cells (PEMFC) specific characteristics. The proposed scheduling process makes use of an optimal approach based on a Mixed Integer Linear Program (MILP). Efficiency of the associated commitment strategy is assessed by comparison with basic intuitive strategies, considering constant and piecewise constant load demand profiles.

In a post-prognostics decision context, this paper addresses the problem of maximizing the useful life of a platform composed of several parallel machines under service constraint. Application on multi-stack fuel cell systems is considered. In order to propose a solution to the insufficient durability of fuel cells, the purpose is to define a commitment strategy by determining at each time the contribution of each fuel cell stack to the global output so as to satisfy the demand as long as possible. A relaxed version of the problem is introduced, which makes it potentially solvable for very large instances. Results based on computational experiments illustrate the efficiency of the new approach, based on the Mirror Prox algorithm, when compared with a simple method of successive projections onto the constraint sets associated with the problem.