As the power industry deregulates and competition are introduced, the scale of market subject is increasing in China. The energy market and auxiliary service market are forming gradually, and competing in the power market has become a trend for generation companies. The pumped-storage unit is a form of power supplies with quick response and flexible operation modes. However, the competitiveness of it is limited due to lack of proper bidding strategy. Therefore, a bi-level bidding strategy is proposed in this paper to quantitatively analyze the optimal bidding prices of a pumped-storage plant with a number of units. The bidding model is built within the rules and constraints of grids, and solved by the Genetic Algorithm. By solving the lower-level model, the minimum generating cost of whole system and appropriate capacity allocation of pumped-storage plant are achieved. And the target of the upper-level model is maximizing the profits of pumped-storage plant. Furthermore, the proposed optimal strategy is compared with the pumped-storage plant cleared-first strategy through a case study. The results of case studies indicate that the optimal bidding strategy proposed in this paper can dispatch the capacity of the pumped-storage plant reasonably, optimize resource allocation and realize the maximum profit of the pumped-storage plant in power market.