The invention relates to the technical field of traffic monitoring and provides a monitoring method based on dynamic traffic flow. The method comprises steps of S1, determining an independent variable parameter and a dependent variable parameter of a training model; S2, recording historical data of the independent variable parameters and the dependent variable parameters; S3, predicting by the BP neural network to obtain a BP predicted value Qbp (t+m) of the downstream traffic flow in one direction of the intersection at the moment t+m, specifically, inputting historical data of respective variable parameters and dependent variable parameters into the BP neural network to obtain a BP neural network traffic model, and inputting numerical values of respective variables into the BP neural network traffic model at the moment t to obtain a BP neural network traffic model; obtaining a BP predicted value Qbp (t+m) of the downstream traffic flow in the direction of the intersection at the moment of t+m; and S4, Q prediction (t+m) being obtained according to Qbp prediction (t+m), and then the monitoring level M of the crossroad in the direction being determined. According to the method, a problem that precision and accuracy of traffic flow prediction still need to be further improved is solved.
本发明涉及交通监控技术领域,提出了一种基于动态交通流量的监测方法,包括S1、确定训练模型的自变量参数和因变量参数;S2、记录以上自变量参数、因变量参数的历史数据;S3、通过BP神经网络预测得到t+m时刻此十字路口其中一方向下游交通流量BP预测值Qbp预测(t+m),具体的,由各自变量参数及因变量参数的历史数据输入至BP神经网络,得到BP神经网络交通模型,t时刻时,各自变量的数值输入至BP神经网络交通模型,得到t+m时刻此十字路口此方向下游交通流量BP预测值Qbp预测(t+m);S4、根据Qbp预测(t+m)获得Q预测(t+m),再决定此十字路口此方向的监控级别M。通过上述技术方案,解决了交通流量预测的精确度和准确度仍然需要进一步的提高问题。
Monitoring method based on dynamic traffic flow
一种基于动态交通流量的监测方法
2021-04-13
Patent
Electronic Resource
Chinese
European Patent Office | 2018
|Road traffic multidirectional dynamic management and control method based on traffic flow monitoring
European Patent Office | 2023
|European Patent Office | 2023
|Traffic flow monitoring method based on monitoring video
European Patent Office | 2021
|