Solid oxide fuel cell anode image segmentation based on a novel quantum-inspired fuzzy clustering

• A novel quantum-inspired fuzzy clustering method is proposed for three-phase identification of SOFC microstructure.
• Quantum-inspired probability distribution is presented to adjust the inaccurate probability estimation of uncertain points.
• The accuracy and effectiveness of three-phase identification on the micro-investigation are improved.

High quality microstructure modeling can optimize the design of fuel cells. For three-phase accurate identification of Solid Oxide Fuel Cell (SOFC) microstructure, this paper proposes a novel image segmentation method on YSZ/Ni anode Optical Microscopic (OM) images. According to Quantum Signal Processing (QSP), the proposed approach exploits a quantum-inspired adaptive fuzziness factor to adaptively estimate the energy function in the fuzzy system based on Markov Random Filed (MRF). Before defuzzification, a quantum-inspired probability distribution based on distance and gray correction is proposed, which can adaptively adjust the inaccurate probability estimation of uncertain points caused by noises and edge points. In this study, the proposed method improves accuracy and effectiveness of three-phase identification on the micro-investigation. It provides firm foundation to investigate the microstructural evolution and its related properties.