Traditional pricing strategies are often based on static analysis and empirical judgment, which makes it difficult to accurately capture market dynamics and uncertainty. How companies can formulate effective pricing strategies to maximize profits has become a key issue. To address this issue, this study innovatively combines the Cournot model of game theory with Deep Reinforcement Learning (DRL) to explore a more intelligent and adaptive pricing strategy optimization method. This paper takes the Cournot model of game theory as the theoretical basis and constructs a framework that reflects the competitive situation in an oligopoly market. On this basis, DQN (Deep Q-Network) is introduced, which regards each enterprise as an intelligent agent to learn and make decisions in a simulated market environment. The intelligent agent continuously adjusts its pricing strategy through interaction with the environment (i.e., the market) to maximize cumulative profits. Compared with the traditional Cournot model, the average revenue growth rate of the pricing strategy combining the Cournot model and DRL proposed in this paper increases from about 5.17% to about 10.81%, the market share increases, and the maximum drawdown is greatly reduced. In addition, the intelligent agent shows excellent dynamic adjustment ability and recovery resilience, and is able to maintain a high profit margin and a low frequency of pricing strategy adjustment in different market environments. These results fully demonstrate the effectiveness of this method in optimizing corporate pricing strategies, improving market competitiveness and risk management capabilities.