Abstract
The proton exchange membrane fuel cell (PEMFC), which is accounted as a reliable fuel cell (FC) technology, is mainly famous for its low operating temperature and expanded lifespan while also being highly efficient. Such merits have turned the PEMFC into an interesting option to be employed in a wide range of applications, comprising those for backup power, portable power, and vehicle power sources. Due to the output power’s dependence on temperature and membrane water content, the PEMFC, however, faces a number of difficulties. The FC system’s maximum output power can only be achieved at one specific operating point under varying operating circumstances. Therefore, maximizing the power obtained from the PEMFC is essential for improved operation and effective exploitation. It is noteworthy that a combinatorial technique using the developed modified manta ray foraging optimization (DMRFO) method and enhanced adaptive neuro-fuzzy inference system (ANFIS)-based incremental conductance (INC) has been deployed in this paper for the maximum power point tracking (MPPT) to optimally increase the power extraction capacity of an FC-resourced energy system. This method consists of two steps. In the first stage, the DMRFO algorithm is used to specify the best power levels while taking temperature and membrane water content fluctuations into account. This information would be utilized as the input–output training dataset in the ANFIS. The INC approach would initialize at this value in the second stage and seek the highest maximum power point (MPP). The dataset with the shrunk size given by the proposed DMRFO method for training the ANFIS in contrast to prevalent ANFIS is one of the benefits of adopting the ANFIS-based INC technique. This approach is also very quick, stable, and efficient. By taking into account a PMFC with a capacity of 7,000 W, the recommended technique is simulated and the results are compared with other methods to validate the performance.