Abstract: By using the power demand response mechanism and leading users to actively participate in the peak load regulation of the power grid, the enthusiasm of users to interact with the power grid will be better stimulated and the power supply pressure during the peak hours will be greatly reduced. For different power users need different pricing packages because of their different power demands, and their decision-making behavior in selecting the pricing packeges is generally boundedly rational rather than fully rational, i.e. such behavior has the features of the evolutionary game process, this paper proposes a customized pricing packages design for the limited rational user choice behavior. First of all, the k-means algorithm is used to cluster the load characteristic indexes of the daily load curve of the power users, and the typical load curve of each type of the users is generated after the clustering. So the power consumption curve of each package is determined according to the typical load curve by using the fuzzy theory. Secondly, the entropy weight method is used to calculate the utility of the users, and the evolutionary game theory is used to calculate the proportion of their package selection in the case of bounded rationality. Then, the model of power purchase contract between the power selling supplying companies is established. Finally, on the premise of guaranteeing the market shares of and the users' satisfaction with the power selling companies, in order to achieve the goal of maximum win-win profits for both the power selling companies and the users, a customized electricity price package model is established for the boundedly rational users. The results of an example prove the rationality and feasibility of the proposed method and model.