Extended black-start can increase the system generation capability for network reconfiguration by starting up more than one generator at the same time. Because of the topological complexity

determination of the energizing paths for an extended black-start scheme is more difficult than that for the conventional one. A new method of searching for the energizing paths for extended black-start with given recovery targets is proposed. An improved method of determining attribute weights by combining the subjective with the objective weight assignment is presented and applied to the attribute weight determination for the energizing paths. The thought of Steiner minimum tree is employed to screen the potential paths first and the grey relation projection method is then employed to make the final decision on the alternative path-set. The effectiveness of the proposed method is validated by the optimization results on a New England 10-unit 39-bus power system and the southern Hebei power system. This work is supported by National Natural Science Foundation of China (No. 51277076) and Fundamental Research Funds for the Central Universities (No. 2014XS75).