Search

To cope with the demand for large amount of flexibility regulation caused by high penetration of intermittent renewable energy, it is necessary to classify and measure the demand capacity for different regulation performance, and to reasonably allocate flexibility resources for different regions and different regulation capacities. This study proposes a flexibility demand analysis and regulation capacity sharing decisions between interconnected power systems considering differences in regulation performance. Firstly, the empirical mode decomposition (EMD) method is used to decompose the historical operating load curves of each sub-region, and the demand capacities of different regulation performances are calculated based on the obtained decomposition results of trend components and fluctuation components. Then, the probability density and the regulation demand capacity interval at different confidence levels are calculated based on the regulation capacity statistics of the sample of historical operation days. Finally, the regulation capacity sharing decisions between the interconnected regions are made based on the cost of various regulation resources in different sub-regions and the confidence level requirements of internal resources in sub-regions to meet regulation demand. A scenario based on the interconnection operation of two regional grids and the self-sufficiency rate of regulation capacity in each sub-region is no less than 0.95 confidence level is used to verify the effectiveness and feasibility of the proposed method. The simulation results demonstrate that the regulation capacity demand considering the difference in regulation quality can provide a detailed basis for the cross-region deployment of different quality flexibility resources, and the total cost of regulation capacity of the regional grid after adopting the cross-region sharing decision model is reduced by about 4.51% compared with the system independent optimization model.

As there are many factors affecting vehicle lane changes in a tunnel, which leads to the unstable state of vehicles during lane changes and an increase of collision events, a new vehicle lane-changing model is proposed by considering the influence of typical factors such as noise and brightness in a tunnel under a vehicle-to-everything (V2X) environment. First, V2X-based technology enables real-time access to a target vehicle surrounding information characteristics in a tunnel, establishing a lane-changing decision model to quantify the willingness of vehicles to change lanes. Second, considering safety as a prerequisite for a lane change, establishing a vehicle safety lane-change-distance model and a minimum safe-distance model was introduced for comparison to evaluate the safety of lane changing. On this basis, considering the noise and brightness effects of a tunnel, the relationship between brightness, noise, and response time of human-driven vehicles, hybrid driving vehicles, and autonomous vehicles is quantitatively analyzed, and then a new vehicle lane-changing model in a tunnel is established. The results of the research show that brightness in a tunnel has a more significant effect on the driver than noise. At the same time, autonomous driving, as well as hybrid driving, has better stability and comfort with less change in velocity, acceleration, and other states during lane changing in a tunnel compared to manned driving, which proves the reasonableness of the model and helps to provide a model basis for research of real vehicle lane changing under a V2X environment. [ABSTRACT FROM AUTHOR]

Accurate spatial-temporal prediction is a fundamental building block of many real-world applications such as traffic scheduling and management, environment policy making, and public safety. This problem is still challenging due to nonlinear, complicated, and dynamic spatial-temporal dependencies. To address these challenges, we propose a novel embedded spatial-temporal network (ESTNet), which extracts efficient features to model the dynamic correlations and then exploits three-dimension convolution to synchronously model the spatial-temporal dependencies. Specifically, we propose multi-range graph convolution networks for extracting multi-scale static features from the fine-grained road network. Meanwhile, dynamic features are extracted from real-time traffic using a gated recurrent unit network. These features can be applied to identify the dynamic and flexible correlations among sensors and make it possible to exploit a three-dimension convolution unit (3DCon) to simultaneously model the spatial-temporal dependencies. Furthermore, we propose a residual network by stacking multiple 3DCon to capture the nonlinear and complicated dependencies. The effectiveness and superiority of ESTNet are verified on two real-world datasets, and experiments show ESTNet outperforms the state-of-the-art with a significant margin. The code and models will be publicly available. [ABSTRACT FROM AUTHOR]

Mobile networks (MNs) are anticipated to provide unprecedented opportunities to enable a new world of connected experiences and radically shift the way people interact with everything. MNs are becoming more and more complex, driven by ever increasing complicated configuration issues and blossoming new service requirements. This complexity poses significant challenges in deployment, management, operation, optimization, and maintenance, since they require complete understanding and cognition of MNs. Artificial intelligence (AI), which deals with the simulation of intelligent behavior in computers, has demonstrated enormous success in many application domains, suggesting its potential in cognizing the state of an MN and making intelligent decisions. In this article, we first propose an AI-powered MN architecture and discuss challenges in terms of cognition complexity, decisions with high-dimensional action space, and self-adaptation to system dynamics. Then potential solutions associated with AI are discussed. Finally, we propose a deep learning approach that directly maps the state of an MN to perceived QoS, integrating cognition with the decision. Our proposed approach helps operators to make more intelligent decisions to guarantee QoS. Meanwhile, the effectiveness and advantages of our proposed approach are demonstrated on a real-world dataset involving 31,261 users over 77 stations within 5 days. [ABSTRACT FROM AUTHOR]

Driver’s emotion is a psychological reaction to environmental stimulus. Driver intention is an internal state of mind, which directs the actions in the next moment during driving. Emotions usually have a strong influence on behavioral intentions. Therefore, emotion is an important factor that should be considered, to accurately identify driver’s intention. This study used the support vector machine (SVM) theory to develop a driver intention recognition model, with the involvement of driver’s emotions. Various materials including visual materials, auditory materials, and olfactory materials, were used to induce driver’s emotions. Real driving, virtual driving and computer simulation experiments were conducted to collect human-vehicle-environment dynamic data in two-lane roads. The results present that the proposed model can achieve high accuracy and reliability in recognizing driver’s intentions. Our findings of this study can be used to develop the personalized driving warning system and intelligent human-machine interaction in vehicles. This study would be of great theoretical significance for improving road traffic safety. [ABSTRACT FROM AUTHOR]