Reinforcement learning (RL)-based energy management is one of the current hot spots hybrid electric vehicles. Recent advances in RL-based focus on energy-saving performance but less considers constrained setting for training safety. This article proposes an RL framework named coach-actor-double-critic (CADC) optimization considered as Markov decision process (CMDP). A bilevel onboard controller includes a neural network (NN)-based strategy actor and rule-based coach online management. Once...

Coordinating a platoon of connected hybrid electric vehicles (HEVs) poses challenges due to the intricacy their powertrains and diverse driving scenarios encountered. The existing mainstream framework uses hierarchical control scheme, simplifying unified optimization problem into two separate series processes: powertrain level vehicle level. However, this approach overlooks inherent interdependence between systems, which can hinder effective collaboration in terms energy management across...

Understanding the underlying patterns of urban mobility dynamics is essential for both traffic state estimation and management facilities services. Due to coupling relationship generative factors in spatial-temporal domain, it challenging model citywide under a structural pattern critical features such as hours days, days weeks weather conditions. To address this challenge, article develops disentangled representation learning framework learn an interpretable factorized independent data...

Cooperative control of connected and automated electric vehicles (CAEVs) offers a great potential for safe, high-efficient, sustainable transportation system. Among them, coordinated on-ramp merging based on an vehicle mainline facilitating is hot spot to lighten the impact shockwave highway junctions. Reinforcement learning (RL) promising solution address this problem its strong adaptiveness self-learning ability. The existing methods are dedicatedly designed circumvent sparse reward...

As we all know that traditional three tubes treatment for the anastomotic fistula after esophagectomy is easy and efficient, but there are still so many problems such as pain, intubation discomfort, psychological pressure prognosis slowly. Our report would introduce a new method to cure patient radical radiotherapy.CT DSA-guided percutaneous cervical gastric wall puncture were performed while Flocare tube was introduced entering mediastinum via under guidance of contrast medium followed by...

Abstract Urban rail transit has become an indispensable mode in major cities worldwide regarding the advantages of large capacity, high speed, punctuality, and environmental protection. Origin‐destination (OD) matrix data is crucial to organisation train operation management. Nevertheless, OD matrices are inevitably suffered from loss problems due transmission acquisition failures. Tensor completion a state‐of‐the‐art method for missing imputation. In this paper, novel tensor OD‐ proposed....

The study of urban road vehicle lane changing behavior can provide effective decisions for the safety management section traffic. To this kind behavior, random utility theory was used to analyze influence factors like traffic volume, vehicles' speed and distance between vehicles, drivers' characteristics behavior. A probability model established based on greatest utility, a genetic algorithm calibrate parameters. Taking roads Xi'an as an example, we conducted observations data collection...

Obtaining high resolution traffic states of large-scale freeway is always a significant topic for both transportation engineers and researchers. This paper presents machine learning based speed estimation method using two data sources. Two low heterogeneous are collected from microscopic simulations with different error distributions. A neural network model implemented fusing the sources improving time space estimations. The validation results sensitivity analysis indicate that proposed...

Environmental noise induced by metro has drawn increasingly high attention among public. In this paper, environmental noises at the platform and inside carriage are detected field test in Xi'an Metro. The results showed intensity exceeded standard (70 dB) listed quality for noise, operation is main cause, followed number volume of public broadcasting, passengers, trains operation. within carriages influenced infrastructural quality, speed, vehicle dynamic characteristics. At end available...

With the continuous development of its economic society, car ownership in Xi'an keeps growing; as a result, there has been strong push for "green travel" to promote "transit trip" and bicycle travel. Based on increased use travel, this paper studies benefits creating bike sharing system Xi'an. This presents residents using point-to-point rental station. Furthermore, bi-level programing model, which minimizes regional travel construction costs implementing provides foundation slow-traffic

Describing the traffic characteristics of pedestrians and analyzing station carrying capacity is important for design planning operation management. Therefore this paper analyzes pedestrian behavior proposes concepts basic inbound capacity. The definition calculation method then proposed. There a large gap between solution model by using analytical methods actual analysis, so simulation algorithm based on "stress test" Anylogic software. Station bottlenecks are found to provide suggestions...