As the global demand for sustainable transportation solutions grows, fuel cell hybrid vehicles (FCHVs) are increasingly recognized due to their low emissions and high efficiencies. However, energy management, particularly the power distribution between fuel cells and batteries, remains significant challenges. This research proposes an energy management strategy (EMS) based on the Proximal Policy Optimization (PPO), a deep reinforcement learning algorithm, that effectively controls the output power of the fuel cell to enhance the energy distribution. The PPO-based EMS, focusing on improving the fuel economy, enables stable training through small policy updates, which helps to optimize the energy management performance and demonstrate robust adaptability. To assess the effectiveness of the proposed PPO-based EMS in terms of fuel economy, comparisons are made with between the PPO-based EMS and the rule-based EMS, with the dynamic programming (DP)-based EMS serving as the benchmark. The results indicate the PPO-based EMS achieves more effective energy distribution and improves fuel economy effectively.