China’s multi-level unified power market system has begun to take shape, and the idea of regional joint clearance has further expanded the scope of optimal allocation of power resources. However, in the process of constructing the new power system, its high proportion of renewable generation characteristics increases the difficulty of power market region clearance. Considering that the uncertainty of both source and load sides is greatly enhanced, an ant colony ranking clustering algorithm based on the optimal cluster number selection is proposed to obtain the characteristics of the uncertainty parameter distribution. On this basis, probabilistic characterization is introduced, and a day-ahead regional transaction clearing model with uncertainty is proposed to help improve the accuracy and applicability of the clearing results, which can minimize the transmission cost and reduce inter-provincial purchasing costs. The simulation results verify the accuracy of the model and the rapidity of the solution.