Search

User preferences are important for electric vehicle charging station (EVCS) recommendations, but they have not been deeply analyzed. Therefore, in this study, user charging preferences are identified and applied to EVCS recommendations using a hybrid model that integrates LightGBM and singular value decomposition (SVD). In the model, LightGBM is used to predict user ratings according to users’ comments regarding charging orders, and the feature importance reported by each user is output. Then, a co-occurrence matrix between users and charging stations (EVCSs) is constructed and decomposed using SVD. Based on the decomposed results, the final evaluated scores of each user for EVCSs can be calculated. Upon ranking the EVCSs according to the scores, the EVCS recommendation results are obtained, taking into account the users’ charging preferences. The sample data consist of 28,306 orders from 508 users at 241 charging stations in Linyi, Shandong, China. The experimental results show that the proposed hybrid model outperforms the benchmark models in terms of precision, recall, and F1 score, and its F1 score can be increased by 96% compared with that of the traditional item-based collaborative filtering method with charging counts for EVCS recommendations.

As the commercialisation of Intelligent Connected Vehicles (ICVs) accelerates, Vehicle-to-Everything (V2X)-based general testing and assessment systems have emerged at the forefront of the research. Current field testing schemes mostly follow the norms of traditional vehicle tests. In contrast, Original Equipment Manufacturers (OEMs) have increasingly focused on the potential influence of V2X communication performance on the application response characteristics. Our previous work resulted in a C-V2X (Cellular-V2X) large-scale testing system (LSTS) for communication performance testing. However, when addressing the need to combine application and communication, the system software faces confronts heightened technical challenges. This paper proposes a layered software architecture for the automated C-V2X LSTS, which is tailored to combined scenarios. This architecture integrates scenario encapsulation technology with a large-scale node array deployment strategy, enabling communication and application testing under diversified scenarios. The experimental results demonstrate the scalability of the system, and a case study of Forward Collision Warning (FCW) validates the effectiveness and reliability of the system.

Bus arrival time prediction not only provides convenience for passengers, but also helps to improve the efficiency of intelligent transportation system. Unfortunately, the low precision of bus-mounted GPS system, lack of real-time traffic information and poor performance of prediction model lead to low estimation accuracy - greatly influence bus service performance. Hence, in this paper, a GPS calibration method is put forward, while projection rules of specific road shapes are discussed. Moreover, two traffic factors, travel factor and dwelling factor, are defined to express real-time traffic state. Then, considering both historic data and real-time traffic condition, a hybrid dynamic BAT prediction factor, which achieves accuracy enhancement by taking into account traffic flow evaluation results and GPS position calibration, is defined. A LSTM training model is construct to realize BAT prediction. Experiment results demonstrate that our technique can provide a higher level of accuracy compared to methods based on traditional time-of-arrival techniques, especially in the accuracy of multi-stops BAT prediction.

With the concept of edge computing being put forward, the edge node should play an important role in resource allocation and local computation. However, there are two basic problems to confront in this arena: the performance of basic safety message (BSM) exchange and the frequency of service migration, both of which are related to the vehicle's mobility feature. To decrease the frequency of service migration and enhance the efficiency of the BSM exchange, in this paper, a bus-oriented mobile fog computing nodes infrastructure is proposed, while a mobile similarity-based clustering method is given. In the proposed method, the bus node, which is considered as a fog-computing node (FCN), is used as a sink node, which fulfills communication resource allocation and convergence computing tasks. MATLAB is used to simulate the path loss feature, while a network simulator (NS-3) is employed to investigate the performance of the proposed scheme. The simulation results show that the proposed scheme significantly decreases the frequency of service migration and perform high communication efficiency for intra-cluster BSMs exchange.

Private cars are the most active and important incremental factor in the electric vehicle market and are expected to account for 80% of the new energy vehicle sales market by 2030. As the most common charging scenario for private cars, orderly charging in the community can optimize the distribution load curve by dynamically adjusting charging time and power of electric vehicles, so as to achieve peak-load shaving and turn electric vehicles into a friendly load to the distribution grid. However, after the traditional orderly charging strategy was released, the complexity of the operation on the user’s side was a heavy strike to the user’s willingness to participate in orderly charging, resulting in the quite low participation rate and insufficient demonstration of the characteristics of EV’s elastic power demand. To solve this issue, the paper proposes a senseless orderly charging strategy with user charging demand prediction and substation capacity constraint considered to minimize user charging fee or maximize service provider revenue. After a five-month practical application, the proposed strategy was found to effectively improve the user participation rate in orderly charging and to regulate electric vehicles as an elastic load to meet the grid demand.

Fatigued driving detection, which employs pattern recognition to discover the state of a driver's fatigue, is considered a key technique to improve road safety. However, the widely used frontal face recognition systems have problems, such as low recognition accuracy, poor real-time ability, and highly complex algorithms. In this paper, a new color space modeling, which uses multi-threshold decision criteria, is used to enhance facial skin extraction performance, and a three-step strategy is designed to eliminate noise and other adverse effects. Experiments show that the proposed algorithm can extract side face contour lines effectively and provide a scientific basis for real-time tracking of fatigue.