Recent years have witnessed the flourish of new promising technologies to prolong the lifetime of wireless sensor networks (WSNs). Employing mobile vehicles for wireless charging and data collection is able to balance the load of WSNs, and meanwhile, provide a reliable energy replenishment for sustainable network operations. Different from traditional charging scheduling policies where sensor nodes passively wait for the arrival of mobile vehicles, a novel dynamic clustering based mobile-to-cluster (M2C) scheme is proposed to optimize the service process for both sensor nodes and the vehicle in an active way. Requiring only local residual energy information, M2C refreshes the cluster-based network topology to ensure no sensor nodes will run out of energy before getting charged. By estimating the energy consumption of being a cluster head, the sensor nodes with residual energy close to the estimation are actively elected as head nodes. More head nodes with low residual energy will be visited, reducing the travel distance along with a higher energy efficiency of charging. Furthermore, the convergence and worst-case performance of M2C are theoretically analyzed. Performance evaluation results show that compared with the state-of-the-art schemes, our design reduces the travel distance and charging latency by about 10% and 50%, respectively.

As the two classical power allocation methods in battery-supercapacitor hybrid energy storage systems, split-frequency methods and power-level methods have been developed separately for many years. In this article, we made an attempt to integrate the advantages of the two methods and proposed an adaptive frequency-split-based quantitative power allocation strategy. First, an adaptive power preallocation is quantitatively determined according to the state of charge (SoC) of batteries and supercapacitors. Then, a windowed fast Fourier transform (FFT)-based power spectrum calculation algorithm is designed to derive the power level corresponding to each sampling frequency. By mapping the preallocated power to the power spectrum, the split frequency is adaptively computed. Finally, the power allocation is implemented through a low-pass filter (LPF) with the derived split frequency. Extensive experiments verify that the proposed method provides an improved performance in suppressing the dc bus voltage fluctuations and protecting batteries when compared with existing methods.

Driving style recognition plays a key role in ensuring driving safety and improving vehicle traffic efficiency. With the development of sensing technology, data-driven methods are more widely uesd to recognize driving style. However, adequately labeling data is difficult for supervised learning methods, while the classification accuracy is not sufficiently approved for unsupervised learning methods. This paper proposes a new driving style recognition method based on Tri-CatBoost, which takes CatBoost as base classifier and effectively utilizes the semi-supervised learning mechanism to reduce the dependency on data labels and improve the recognition ability. First, statistical features were extracted from the velocity, acceleration and jerk signals to fully characterize the driving style. The kernel principal component analysis was used to perform nonlinear feature dimension reduction to eliminate feature coupling. CatBoost is an ensemble of symmetric decision trees whose symmetry structure endows it fewer parameters, faster training and testing, and a higher accuracy. Then, a Tri-Training strategy is employed to integrate the base CatBoost classifiers and fully exploit the unlabeled data to generate pseudo-labels, by which the base CatBoost classifiers are optimized. To verify the effectiveness of the proposed method, a large number of experiments are performed on the UAH DriveSet. When the labeling ratio is 50%, the macro precision of Tri-CatBoost is 0.721, which is 15.7% higher than that of unsupervised K-means, 1.6% higher than that of supervised GBDT, 3.7% higher than that of Self-Training, 0.7% higher than that of Co-training, 1.5% higher than that of random forest, 6.7% higher than that of decision tree, and 4.0% higher than that of multilayer perceptron. The macro recall of Tri-CatBoost is 0.744, which is also higher than other methods. The experimental results fully demonstrate the superiority of this work in reducing label dependency and improving recognition performance, which indicates that the proposed method has broad application prospects.

For periodical beacon broadcasting in cellular vehicle-to-everything (C-V2X) networks, a distributed reservation media access control (MAC) protocol, the sensing-based semi-persistent scheduling (SPS), is adopted. However, how to quantify the communication reliability and latency is an open issue, which is critical for low-latency and high-reliability services. In this paper, an analytical model for SPS is presented, based on which the impacts of beacon rate, range settings and system configuration on access collision probability and delay outage probability are quantified. The analytical model provides important insights and guideline to adapt and optimize protocol parameters including the sensing range, transmit power and resource reservation. The enhanced MAC protocol can maintain high-reliability and low-latency services with a wide range of vehicle density. Simulations are conducted to validate the analysis and the results demonstrate that the proposed MAC enhancement solution can reduce collision probability while ensuring the delay outage probability based on service requirements.

The constant-current constant-voltage (CC-CV) charging method is a common approach to charge batteries with a single charger. Recently, with the emergence of high power-density batteries, e.g., supercapacitors, several chargers are typically connected in parallel to provide the required charging power in many applications. However, the classical decentralized control scheme for the multicharger system leads to the current imbalance, which deteriorates the reliability of the charging system. In this article, we extend the CC-CV charging protocol to multicharger systems using a cooperative control method, which can alleviate the current imbalance among chargers effectively. First, the motivation and charging system modeling of this work are provided. Then, a cooperative CC charging protocol and a cooperative CV charging protocol are presented, respectively, and a switching logic is designed to achieve the switching between them based on the supercapacitor voltage. The closed-loop model of the proposed charging system is developed using the block diagram. A laboratory testbed is built to verify the effectiveness of the proposed method. Experiment results show that the proposed method provides a better current balancing and voltage regulation performance when compared with existing decentralized control methods.

Catenary-free trams with an onboard supercapacitor have been an emerging and energy-efficient way for urban rail transportation. However, the high-power charging load from catenary-free trams brings a new challenge to the design of charging systems. The traditional single-module charging system typically suffers from a high power and thermal pressure, which reduces the reliability of the charging system. In this paper, we propose a multimodule charging system to improve the system reliability by decomposing power pressure to multiple charging modules. Specifically, a cooperative charging protocol is proposed to improve the dynamic performance of the charging system by suppressing the current imbalance and avoiding the current overshoot during the startup process. First, we mathematically model the multimodule charging system using the state-space averaging method and graph theory. Second, the cooperative charging protocol is designed using the leader-follower consensus algorithm, where the virtual leader represents the desired charging current for each module with a warm-start mechanism. Third, multiple-input multiple-output root locus technique is exploited to characterize the dynamic performance of the charging system. Both simulation and experiment results show that the proposed protocol effectively suppresses the imbalance and avoids the overshoot during the startup of the charging system.