Abstract: Under the background of the energy transition and informatization construction, the multi-modal data for the power system exist objectively. The multi-modal data fusion is an effective way to implement the collaborative analysis of multiple heterogeneous networks and to push forward observable and controllable power grids. On this basis, the multi-modal data and multi-modal data fusion for power systems are analyzed. And the multi-modal data fusion frameworks are given including the intranetwork fusion and inter-network fusion, and their respective fusion modes are also divided. Then, with the current research status, the key problems of intra-network fusion and inter-network fusion are summarized, and the corresponding technical solutions are proposed. Finally, from the aspects of the concept, fusion mode, key technical issues, and solutions, the multi-modal data fusion for power systems is summarized.