This paper presents the design of a coordinated ramp metering strategy based on a newly developed multi-class second-order macroscopic traffic flow model coupled with a pollutant emissions model. The approach followed for obtaining the parameters of the multi-class model is based on sigmoid surfaces spanning over traffic density and composition. An optimal control problem is formulated for designing an environmentally sustainable, efficient and equitable coordinated ramp metering strategy. In addition to the global network-wide objective of total emissions minimisation, two different types of environmental constraints are introduced for addressing local requirements. The resulting static large-scale optimisation problem is solved using the Differential Evolution algorithm. A small case study is provided analysing the multi-class flow model and the strategy’s application.

A varying parameter multi-class second-order macroscopic traffic flow model for coordinated ramp metering with global and local environmental objectives

The accurate online soft sensor in complex industrial processes remains challenging because underlying spatial coupling relations among process variables have not been effectively mined and exploited. Recently, some deep learning based studies construct static graphs to explicitly represent underlying spatial coupling relations among process variables, but they neglect the fact that spatial coupling relations have mutable characteristics, leading to poor performance in the online soft sensor. Therefore, in this article, we propose a novel deep learning model to address this issue to achieve accurate soft sensors. Specifically, a dynamic graph is proposed to realize adaptive learning and automatic inference for mutable spatial coupling relations so that the proposed model is endowed with the ability to real-timely sense spatial coupling relations in the online industrial soft sensor. Then, based on the dynamic graph, a new multihop attention graph convolutional network is proposed to systematically aggregate various crucial node feature representations during graph convolution processes to capture fine-grained spatial dependence features, thereby achieving effective modeling for variation patterns of process variables. Finally, a new multivariate incremental training algorithm is designed for deep learning models to further improve the prediction performance. The verification study on a coal mill rig demonstrates the feasibility and effectiveness of the proposed model.

Dynamic Graph-Based Adaptive Learning for Online Industrial Soft Sensor With Mutable Spatial Coupling Relations

Accurate online soft sensor in complex industrial processes remains challenging because underlying spatial coupling relations among process variables have not been effectively mined and exploited. Recently, some deep learning-based studies construct static graphs to explicitly represent underlying spatial coupling relations among process variables, but they neglect the fact that spatial coupling relations have mutable characteristics, leading to the poor performance in online soft sensor. Therefore, we propose a novel deep learning model to address this issue to achieve accurate soft sensor. Specifically, a dynamic graph is proposed to realize adaptive learning and automatic inference for mutable spatial coupling relations, so that the proposed model is endowed with the ability to real-timely sense spatial coupling relations in online industrial soft sensor. Then, based on the dynamic graph, a new multi-hop attention graph convolutional network is proposed to systematically aggregate various crucial node feature representations during graph convolution processes to capture fine-grained spatial dependence features, thereby achieving effective modeling for variation patterns of process variables. Finally, a new multivariate incremental training algorithm is designed for deep learning models to further improve the prediction performance. Verification study on a coal mill rig demonstrates the feasibility and effectiveness of the proposed model. 

We propose a Bayesian approach for parameter uncertainty quantification in macroscopic traffic flow models from cross-sectional data. We consider both a simple first order model consisting in the mass conservation equation and its second order version including a speed evolution equation. A bias term is introduced and modeled as a Gaussian process to account for the traffic flow models limitations. We validate the results comparing the error in the macroscopic variables (flow, speed, density) for both models, showing that second order models globally perform better in reconstructing traffic quantities of interest. 

/pdf/data-driven-uncertainty-quantification-in-macroscopic-3uyuqh1s.pdf

Data-driven uncertainty quantification in macroscopic traffic flow models

Traffic state estimation (TSE) is a critical component of the efficient intelligent transportation systems (ITS) operations. In the literature, TSE methods are divided into model-driven methods and data-driven methods. Each approach has its limitations. The physics information-based neural network (PINN) framework emerges to mitigate the limitations of the traditional TSE methods, while the state-of-art of such a framework has focused on single road segments but can hardly deal with traffic networks. This paper introduces a PINN framework that can effectively make use of a small amount of observational speed data to obtain high-quality TSEs for a traffic network. Both model-driven and data-driven components are incorporated into PINNs to combine the advantages of both approaches and to overcome their disadvantages. Simulation data of simple traffic networks are used for studying the highway network TSE. This paper demonstrates how to solve the popular LWR physical traffic flow model with a PINN for a traffic network. Experimental results confirm that the proposed approach is promising for estimating network traffic accurately.

https://www.mdpi.com/1999-4893/15/12/447/pdf?version=1669890084

Physics-Informed Neural Networks (PINNs)-Based Traffic State Estimation: An Application to Traffic Network

Traffic data missing issues due to unpredictable equipment failure, extreme weather, and other reasons have brought great challenges to traffic flow prediction modeling. In this paper, a novel reinforced dynamic graph convolutional network model is proposed to simultaneously conduct data imputation and network-wide traffic flow prediction. First, a multi-graph convolutional fusion network is proposed for data imputation by using the graph convolutional network to analyze the propagation law of traffic states between traffic flow detection stations in both time and space dimensions. Second, to enhance the robustness of network-wide traffic flow prediction, a dynamic graph learning method based on deep reinforcement learning is proposed to adaptively generate the graph adjacency matrix to represent the dynamic spatiotemporal dependencies between the stations. Finally, experimental results on two real-world traffic datasets show that the proposed method outperforms other baseline methods and can effectively extract the data missing features and spatiotemporal dependence features between the stations. The visualization results of the graph adjacency matrix indicate that the proposed method can effectively identify the influential traffic stations in the process of traffic flow prediction, and the extracted dependencies between the stations are interpretable. The proposed model has strong generalization in tackling network-wide traffic flow prediction tasks with different data missing rates and missing patterns, and can be extended to assist decision-makers in enhancing traffic management and mitigating traffic congestion. 

A novel reinforced dynamic graph convolutional network model with data imputation for network-wide traffic flow prediction

This paper uses empirical traffic data collected from three locations in Europe and the US to reveal a three-phase fundamental diagram with two phases located in the uncongested regime Model-based clustering, hypothesis testing, and regression analyses are applied to the speed-flow-occupancy relationship represented in the three-dimensional space to rigorously validate the three phases and identify their gaps The finding is consistent across the aforementioned different geographical locations Accordingly , we propose a three-phase macroscopic traffic flow model and a characterization of solutions to the Riemann problems This work identifies critical structures in the fundamental diagram that are typically ignored in first-and higher-order models, and could significantly impact travel time estimation on highways

https://hal.inria.fr/hal-01864628/document

A three-phase fundamental diagram from three-dimensional traffic data

Bibliometric methods in traffic flow prediction based on artificial intelligence

Information systems record the current states and the access records in logs, so logs become the data basis for detecting anomalies of system security. To realize log anomaly detection, frameworks based on text, sequence, and graph are applied. However, the existing frameworks could not extract the complex associations in logs, which leads to low accuracy. To meet the requirements of the hyperautomation framework for log analysis, this paper proposes GenGLAD, a generated graph based log anomaly detection framework. The generated graph is used to express the log associations, and the node embedding of the generated graph is obtained based on random walk and word2vec. Finally, we use clustering to realize unsupervised anomaly detection. Experiments verify the detection effect of GenGLAD. Compared with the existing detection frameworks, GenGLAD achieves the highest accuracy and improves the comprehensive detection effect. 

Yong Chen

Papers

A novel reinforced dynamic graph convolutional network model with data imputation for network-wide traffic flow prediction

A three-phase fundamental diagram from three-dimensional traffic data

Bibliometric methods in traffic flow prediction based on artificial intelligence

GenGLAD: A Generated Graph Based Log Anomaly Detection Framework

Understanding common human driving semantics for autonomous vehicles