高速公路收费站车流量短期预测数据

该数据以高速公路收费站为载体，以收费站、门架历史数据为训练集，结合收费站、门架实时数据，利用改进后的STGCN时空卷积网络算法，构建收费站车流量短期预测模型，实时预测未来每5分钟、10分钟、30分钟的收费站车流量情况。经验证，预测准确度可达95%以上。实时准确的高速公路短时流量预测是高速公路智慧运营的重要内容，有利于实现高速运营管理者合理有效配置收费人员、采取动态交通诱导、提前开启收费站潮汐车道，有效缓解收费站高峰时期的拥堵问题、实现用数据辅助决策，达到智慧化交通运营管理的成效。

This dataset takes expressway toll stations as the core application scenario, uses historical data from toll stations and traffic gantries as the training dataset, and combines with real-time data from the same toll stations and gantries. An improved Spatial-Temporal Graph Convolutional Network (STGCN) algorithm is employed to construct a short-term traffic flow prediction model for toll stations, which can perform real-time prediction of toll station traffic volumes at 5-minute, 10-minute, and 30-minute forecasting horizons. Empirical validation demonstrates that the prediction accuracy exceeds 95%. Real-time and accurate short-term expressway traffic flow prediction is a critical component of intelligent expressway operations. It enables expressway operation managers to rationally and efficiently allocate toll station staff, implement dynamic traffic guidance, open tidal lanes at toll stations in advance, effectively mitigate peak-hour congestion at toll stations, realize data-driven decision-making, and achieve the objectives of intelligent traffic operation and management.