An improved MPPT control-based ANFIS method to maximize power tracking of PEM fuel cell system. (December 2022)

• Record Type: Journal Article
• Authors: Hai, Tao; Wang, Dan; Muranaka, Tetsuya
• Is Part Of: Sustainable energy technologies and assessments. Volume 54(2022)
• Publication Date: 2022-12
• Subjects: PEMFC System, MFSO algorithm -- Incremental conductance -- MPPT controller; Renewable energy sources -- Periodicals; Energy development -- Technological innovations -- Periodicals; Electric power production -- Periodicals; Energy storage -- Periodicals; 333.79
• DOI: 10.1016/j.seta.2022.102629
• Languages: English
• ISSNs: 2213-1388

Abstract

Abstract: The main benefits of proton exchange membrane fuel cell (PEMFC) Include high efficiency, long lifetime and low operating temperature. Due to the mentioned benefits, the PEM fuel cell is used widely in applications such as buses, recreation vehicles, and lightweight vehicles. Nevertheless, the main limitations of them Include the high dependency of their output power on the temperature of cell and membrane water content. There exist one unique operating point for such a system with maximum output under variation operation circumstances. Consequently, it is required for the PEM fuel cell to extract the maximum power to improve efficiency. Accordingly, a promoted MPPT strategy is designed in this paper deploying the Incremental conductance (Inc) method optimized by employing a modified fluid search optimization algorithm (MFSO) algorithm-based adaptive neuro-fuzzy inference system (ANFIS). This technique Includes two stages specifying the optimal values of power by deploying the MFSO algorithm taking into consideration variations of temperature and membrane water content used as the input–output training information set in the ANFIS in the first stage. The second stage, the Inc technique would initialize at this value and seek the maximum power point (MPP). One of the merits associated with using the ANFIS-based Inc technique is the smaller dataset needed for the training purpose. In this regard, in case the ANFIS does not specify the definite point, the Inc approach would do it. The proposed method is simulated, and the results derived verify that the MFSO-ANFIS-based Inc MPPT algorithm would efficiently lead to the global maxima taking into account various climate conditions. This method is also highly efficient, fast, and stable. The efficiency of the suggested method is properly investigated by considering a 7 kW PEMFC fed by a DC-DC boost converter in Matlab/Simulink environment.