Aiming at the major emergency decision-making problem when the weights of decision-makers are unknown, a SFN-CFSFDP-Borda-MULTIMOORA method is proposed. Firstly, spherical fuzzy number is used to describe the evaluation information of experts on emergency alternatives. Secondly, the clustering method based on the theory of clustering by fast search and find of density peaks is used to cluster the large group experts. Then, according to the clustering results, the conflict is measured and an extensible multi-objective optimization model of group comprehensive conflict is constructed, the weights of experts and clusters are obtained and group conflicts are effectively reduced by solving the model. Thirdly, the optimal alternative is decided by the Borda-MULTIMOORA method. Finally, taking Donghu reservoir in Heilongjiang province as an example, the effectiveness and practicability of the method are verified. The results show that the proposed method fully considers the psychological characteristics of experts, strengthens the knowledge expression ability of experts in major emergency scenarios, and clarifies the selection method of clustering centers, which achieves better clustering effect with small internal differences and large differences among clusters. Besides, it is more practical by considering the comprehensive conflicts of groups and actual decision-making situations to determine the weights of experts and clusters, which effectively reduces group conflicts, improves the efficiency of decision-making and provides support for the large group emergency decision-making of major emergencies.