Dynamic Programming Algorithm for minimizing operating cost of a PEM fuel cell vehicle

A PEM (Proton Exchange Membrane) fuel cell city bus utilizes a PEM fuel cell engine as the primary source, and a li-ion battery system as the auxiliary power source. By optimizing the power split strategy and recycling braking energy, this kind of power-train has advantages of zero emission and high energy efficiency. However, the cost of hydrogen gas is far more expensive than that of the electric energy. How to split the power between the two power sources so as to minimize the operating cost, as well as guarantee the vehicle dynamic performance, becomes an important topic. This paper proposes a Dynamic Programming Algorithm (DPA) to solve the minimizing problem. Some details of the DPA are discussed, e.g. the principles of selecting parameters for the algorithm. The effectiveness of the algorithm is verified by comparing simulating results of different algorithms. Results show that, 1) by using the DPA algorithm, we can find the optimal control strategy in an objective way. 2) The constraints of vehicle dynamic performance on the optimal problem have great influences on the optimal results. 3) To predict the power requirement in the near future is very important to achieve an optimal real-time strategy.