基于模态空间分析的GIS振动源定位和动载荷重构方法OA

To address the difficulty of quantitatively identifying internal mechanical defects in gas-insulated switch-gear(GIS)through external vibration signals,a study on vibration source localization and dynamic load reconstruction for three-dimensional thin enclosures is performed.A joint inversion framework centered on modal space analysis is constructed,in which a physics-informed Gaussian process(PIGP)combined with a finite-element-mode-based spa-tial shape dictionary is used to perform constrained reconstruction of modal forces.In the vibration source optimiza-tion,a physical constraint based on modal energy distribution is introduced and integrated with a genetic algorithm(GA)to achieve global optimization of the load application position.Based on a GIS enclosure finite element model and a full-scale experimental platform,the acceleration response data is acquired under different excitation paths,background vibrations,and noise conditions,and the localization accuracy as well as the reconstructed load time his-tories are comparatively analyzed.The results show that the proposed joint inversion method can still stably identify internal excitation locations and their dominant temporal characteristics under sparse measurement points and multi-source interference,providing an effective technical approach for physical tracing and quantitative evaluation of GIS mechanical defects.