The invention relates to a multi-graph convolution traffic propagation flow prediction method and system considering periodic volatility, and belongs to the technical field of traffic prediction. The method comprises the following steps: 1, identifying traffic propagation flow information, and constructing a flow propagation quantity matrix; 2, constructing an adjacent propagation graph, a homologous propagation graph and a same-destination propagation graph by using road network information and historical data; thirdly, mining the periodical volatility of the traffic propagation flow based on an input time sequence; fourthly, constructing a traffic propagation flow prediction model of multi-graph convolution considering periodic volatility; 5, training and evaluating the model; the system comprises a matrix construction module, a propagation graph construction module, a time feature mining module, a model construction module and a training and evaluation module. The influence of time periodicity and volatility and the influence of multi-space correlation characteristics are fully considered.
本发明涉及一种考虑周期波动性的多图卷积交通传播流预测方法及系统,属于交通预测技术领域;方法包含第一,对交通传播流信息进行识别,构建流量传播量矩阵;第二,利用路网信息以及历史数据,构建邻接传播图、同源传播图、同目的地传播图;第三,基于输入时间序列挖掘交通传播流周期性的波动性;第四,构建考虑周期波动性的多图卷积的交通传播流预测模型;第五,对模型进行训练和评估;系统包括矩阵构建模块,传播图构建模块,时间特征挖掘模块,模型构建模块,训练和评估模块;充分考虑时间周期性和波动性的影响,以及多空间关联特征的影响。
Multi-graph convolution traffic propagation flow prediction method and system considering periodic volatility
考虑周期波动性的多图卷积交通传播流预测方法及系统
2023-10-17
Patent
Electronic Resource
Chinese
Periodic adaptive graph convolution circulation network traffic flow prediction method
European Patent Office | 2023
|Transform and multi-graph fusion convolution-based traffic flow prediction method
European Patent Office | 2024
|Traffic flow prediction method based on double graph convolution
European Patent Office | 2024
|Double-end graph convolution traffic flow prediction method with graph learning
European Patent Office | 2023
|Dynamic graph convolution traffic speed prediction method
European Patent Office | 2020
|