To address the challenges of spatial localization ambiguity and insufficient temperature rise trend analysis in thermal fault detection of transmission equipment, this paper proposes a multi-modal point cloud and thermal imaging fusion method. High-precision registration between 3D point clouds and thermal imaging data is achieved through camera calibration parameters, and a 4D thermal field reconstruction model is constructed by integrating temperature normalization techniques. Experiments on simulated transmission platforms involving insulators, power lines, and composite systems demonstrate that the fusion model achieves an average accuracy exceeding 92.6%. This method overcomes the spatial limitations of traditional 2D thermal imaging, providing a new paradigm for thermal fault diagnosis of transmission equipment that combines high geometric precision with dynamic monitoring capabilities, thereby significantly enhancing the reliability of smart grid condition-based maintenance.