论文部分内容阅读
为保证城市快速路段的畅通,建立一种基于蜂群算法-支持向量机(ABC-SVM)融合多源交通数据的城市快速路交通事件检测方法。首先通过分析路段实际情况,选取不同检测器的数据作为事件检测模型的输入值;然后利用蜂群算法(ABC)对支持向量机(SVM)分类模型中的参数进行优化,获得最优的交通事件检测模型,模型的输出结果为检测路段是否发生交通事件;最后结合成都市三环城市快速路路段上采集到的多源交通数据进行实例验证。结果表明,利用ABC-SVM方法进行事件检测的效果优于BP神经网络的方法。
In order to ensure the smooth flow of urban expressways, an urban rapid transit traffic incident detection method based on bee colony algorithm-support vector machine (ABC-SVM) fusion of multi-source traffic data is established. Firstly, the data of different detectors are selected as the input value of the event detection model by analyzing the actual situation of the road section. Then, the parameters in the SVM classification model are optimized by the bee colony algorithm (ABC) to obtain the optimal traffic events The result of the model is to detect whether there is a traffic accident in the road section. Finally, an example is given to verify the multi-source traffic data collected from the sections of expressway in the Third Ring Road of Chengdu City. The results show that using ABC-SVM method for event detection is better than BP neural network.