Smart Cities

The anomaly detection of urban rail vehicle door systems is crucial for the security and reliability of rail vehicles, and unsupervised methods have received particular attention in anomaly detection due to the scarcity of anomaly data. However, existing studies have the following limitations: (1) The dynamic interrelationships between different sensor variables are not explicitly considered, which is detrimental to the learning of normal patterns. (2) The detection performance is limited due to the reliance on manual threshold settings and the lack of confidence measures for different sensor variables. To overcome the abovementioned limitations, a dynamic graph attention-based unsupervised bi-learning framework for the anomaly detection of rail vehicle door systems is proposed in this paper. First, a dynamic graph attention block (DGAT block) is designed to model dynamic interactions between sensor variables and fuse information during signal coding. Then, a bi-learning framework is developed to automatically learn thresholds and quantify the confidence of sensor variables for anomaly detection. The proposed method is verified on a real rail vehicle door system. Compared with the state-of-the-art autoencoder-based methods that are applicable to multivariate time series data, the proposed method improves the best competing precision and F1-score by 7.7% and 4.3%, respectively, while reducing the false positive rate (FPR) by 88.9%.

This study presents a practical framework combining Simulation of Urban MObility (SUMO) with its Traffic Control Interface (TraCI) to evaluate how predefined longitudinal and lateral maneuvers imposed on a single vehicle affect target-vehicle performance and aggregate traffic conditions in a simplified signalized urban network. A baseline scenario was compared with a controlled scenario in which one existing vehicle performed distance-triggered speed changes, sub-lane displacements, and a temporary stop. Both scenarios used the same network, traffic demand, routes, departure schedule, simulation settings, and random seed. External control increased the target vehicle’s travel duration from 101.05 to 107.20 s, waiting time from 8.20 to 10.05 s, and time loss from 16.82 to 22.97 s. Network-level differences were substantially smaller: mean network speed decreased from 10.759 to 10.751 m/s, mean completed-trip duration increased from 89.185 to 89.253 s, and 268 vehicles completed their routes in both scenarios. One emergency-braking event was recorded in each scenario, while no collisions or teleportations occurred. The results demonstrate a measurable direct effect on the controlled vehicle, while aggregate network responses remained small and mixed in direction.

Smart-city water management increasingly depends on converting routine sensing data into reliable, actionable models of network assets. Positional control valves are central to pressure regulation, flow allocation, and operational response, yet their actual hydraulic characteristics may differ from manufacturer curves because of aging, maintenance, actuator adjustment, installation conditions, and long-term wear. Inaccurate valve representations can therefore limit the fidelity of hydraulic models, digital twins, and automated control applications. This study proposes in-situ valve characterization using measurements already collected by water utilities. The proposed method estimates the valve characteristic from routine SCADA records of opening, flow, and pressure rather than requiring a dedicated field test. This reduces operational disruption and makes repeated calibration feasible as valve behavior changes over time. It also provides a practical mechanism for transforming existing SCADA archives into updated asset information for model maintenance and condition assessment. The proposed method is robust to common measurement noise and outliers in utility data and is solved efficiently using a least-absolute-residuals formulation implemented via linear programming. The resulting workflow is demonstrated on a real-world case study showcasing a practical path to keep valve curves up to date and improve the fidelity of hydraulic models used for smart-city water network operation and management.