主催: 一般社団法人 日本機械学会
会議名: 2024年度 年次大会
開催日: 2024/09/08 - 2024/09/11
Contamination of metallic foreign particles in membrane electrode assembly (MEA) reduces the performance of the fuel cell used as in power generation devices. In the production line of the MEA, X-ray inspection has been employed to detect the contamination, but it has problems in the cost and the inspection speed. To address these problems, we have developed an automatic inspection system using deep learning into PFPIA-LDS. PFPIA-LDS is a system combining laser displacement sensors (LDS) and a pore fluid pressure impact generator (PFPIG) consisted of a piezoelectric diaphragm and permanent magnets. In PFPA-LDS, an MEA is placed in a static magnetic field and vibrated by the PFPIG. The vibration of MEA elicits eddy currents, and the magnetic force and the Lorentz force due to the interaction between the eddy current and the static magnetic field change the oscillation properties according to the contamination of metallic foreign bodies. In this study, we investigate the factors affecting the oscillation change and the system design of PFPA-LDS through numerical experiments based on the finite element method.