This paper proposes a novel framework for intelligent energy infrastructure planning for electric heavy trucks (EHTs) using vehicle trajectory big data. With EHT sales in China reaching 181,000 vehicles units from January to June 2023 (a 57.7% increase) and supported by national policies, the demand for efficient energy infrastructure is critical. Leveraging GPS trajectory data, traffic flow, and economic factors, the study develops a three-pronged approach: (1) a scenario-based energy replenishment model with a construction evaluation index system, (2) a spatial distribution plan across China’s transportation network, and (3) a real-time intelligent operation management system. Key models, including a demand generation model and an energy consumption rate model, are validated through simulation on Chongqing highway data, achieving a 19% increase in coverage rate (to 91%), a 23% rise in utilization rate (to 88%), and 20% cost savings compared to traditional methods. The Root Mean Square Error (RMSE) improved to 0.12, outperforming alternatives like LSTM and Random Forest. The findings address gaps in dynamic EHT site selection, offering a scalable solution for green logistics. Future research will explore multi-modal integration and real-time data enhancements, with implications for policy and global EV infrastructure.