Data-Driven Detection of Anomalies and Cascading Failures in Traffic Networks
Data-Driven Detection of Anomalies and Cascading Failures in Traffic Networks
复制标题
DOI：
10.36001/phmconf.2019.v11i1.861
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Sanchita Basak;Afiya Ayman;Aron Laszka;A. Dubey;Bruno P. Leao
中科院分区：
文献类型：
--
作者：
Sanchita Basak;Afiya Ayman;Aron Laszka;A. Dubey;Bruno P. Leao
交通网络是任何社区最重要的基础设施之一。智能和互联传感器在交通网络中的日益集成为研究人员提供了研究这一关键社区基础设施动态的独特机会。本文的重点是交通网络的失效动力学。我们特别感兴趣的是分析由物理事件引起的交通拥堵的级联效应，重点是开发隔离和识别拥堵源的机制。为了分析故障传播，开发（a）可以识别异常的监视器和（B）捕获异常传播动态的模型至关重要。在本文中，我们使用真实的交通数据从纳什维尔，TN展示了一种新的异常检测器和定时故障传播图为基础的诊断机制。我们的新颖之处在于能够捕捉的空间信息和交通网络的互连，以及使用递归神经网络架构来学习和预测的操作图形边缘作为其直接的同行，包括传入和传出分支的功能。为了研究物理交通事件，我们增加了真实的数据与使用SUMO，微观交通模拟器生成的模拟数据。我们的结果表明，我们能够构建基于LSTM的交通速度预测器，平均损失为6.55 × 10 ^-4，而基于高斯过程回归的预测器的平均损失为1.78 × 10 ^-2。我们还能够以高精度和高召回率检测异常，导致精确度-召回率曲线的AUC为0.8507。最后，将级联传播问题表示为一个时间故障传播图，可以准确地识别故障源。
Traffic networks are one of the most critical infrastructures for any community. The increasing integration of smart and connected sensors in traffic networks provides researchers with unique opportunities to study the dynamics of this critical community infrastructure. Our focus in this paper is on the failure dynamics of traffic networks. We are specifically interested in analyzing the cascade effects of traffic congestions caused by physical incidents, focusing on developing mechanisms to isolate and identify the source of a congestion. To analyze failure propagation, it is crucial to develop (a) monitors that can identify an anomaly and (b) a model to capture the dynamics of anomaly propagation. In this paper, we use real traffic data from Nashville, TN to demonstrate a novel anomaly detector and a Timed Failure Propagation Graph based diagnostics mechanism. Our novelty lies in the ability to capture the the spatial information and the interconnections of the traffic network as well as the use of recurrent neural network architectures to learn and predict the operation of a graph edge as a function of its immediate peers, including both incoming and outgoing branches. To study physical traffic incidents, we augment the real data with simulated data generated using SUMO, a microscopic traffic simulator. Our results show that we are able to build LSTM-based traffic-speed predictors with an average loss of 6.55 × 10^−4 compared to Gaussian Process Regression based predictors with an average loss of 1.78 × 10^−2. We are also able to detect anomalies with high precision and recall, resulting in an AUC of 0.8507 for the precision-recall curve. Finally, formulating the cascade propagation problem as a Timed Failure Propagation Graph, we are able to identify the source of a failure accurately.