Ziyun Shao’s research while affiliated with Guangzhou University and other places

With the growing use of unmanned surface vehicles (USVs) for underwater exploration, efficient wireless charging solutions like inductive power transfer (IPT) are crucial for addressing power limitations. This paper presents a novel IPT system for USVs and introduces a systematic design approach for optimizing magnetic couplers. The proposed design addresses three critical challenges: misalignment tolerance, lightweight construction, and thermal safety, which are intricately linked through a magnetic field. In terms of misalignment, this paper demonstrates that the coil length is a key factor in determining misalignment tolerance. For a lightweight design, replacing the ferrite plate with ferrite bars can significantly reduce the weight of the coupler without causing core saturation. The design is further validated through a two-way coupled electromagnetic–thermal simulation. The results reveal that, with proper thermal management, the system avoids thermal risks in underwater environments compared to air. Finally, a 3 kW prototype is constructed and tested in fresh water, achieving 55 V and 50 A wireless charging at an 85.7% full-load dc-to-dc efficiency, thus confirming the practicality and performance of the design.

The purpose of this article is to propose a value-oriented electrical load forecasting (ELF) approach that aims to minimize load variance by leveraging vehicle-to-grid (V2G) technology. To achieve this, it is critical and urgent to design a loss function that can accurately measure the suboptimality of decisions induced by forecast errors, especially since the commonly used mean squared error (MSE) falls short in this regard. The Lagrange multiplier method and Karush–Kuhn–Tucker conditions are initially used to elucidate the impact of forecast errors on the actual charging power of electric vehicles. Building on this foundation, a differentiable loss function called mean relative magnitude error (MRME) is put forward for value-oriented ELF, which allows parameter updates in the long short-term memory (LSTM) model via the gradient descent method during the training stage. Furthermore, the MRME and MSE loss functions are combined using a weighted sum method to leverage their respective advantages. Numerous case studies have demonstrated that the LSTM model, whether using MRME alone or in combination with MSE, achieves superior and more robust V2G scheduling performance compared to using MSE alone. The weight coefficients for combining MRME and MSE loss functions are also discussed.

An accurate electrical load forecast is essential for the effective implementation of vehicle-to-grid (V2G) technology to achieve optimal electric vehicle (EV) charging decisions, consequently, ensuring the security and stability of power grid. While prevailing evaluation metrics prioritize forecast quality, they often overlook the significant influence a forecast exerts when integrated into the V2G scheduling optimization. In this paper, a reliable metric is proposed for forecasts in the context of V2G scheduling from the perspective of forecast value. Firstly, we conducted meticulously designed experiments to expose the limitations of forecast quality metrics in the context of V2G scheduling, as well as reveal three key findings. Subsequently, to address computational challenges and enhance representativeness of scheduling results, statistical features of EV charging are used to construct the aggregate model of EV fleet. Then, a reliable metric called V2G scheduling value error (V2G-SVE) is proposed to evaluate the degradation rate of scheduling performance as the score for forecasting performance. Finally, extensive case studies provide compelling evidence for the effectiveness and broad applicability of V2G-SVE. Beyond proposing an evaluation metric, this paper also aims to provide valuable insights about potential direction of improvement for future load forecasting technology.

This work proposes an information security vehicleto- grid (V2G) scheduling solution, which combines Federated deep learning with distributed edge computing for V2G operation. In this framework, each charging point is equipped with an intelligent computing module to conduct distributed edge scheduling for the connected electric vehicle (EV), so that not only the computation of inference process is efficient, but also the privacy-preserving of EV users is guaranteed. Besides, the desensitized V2G data of charging points are used to train the deep neural network model in each charging station. Therefore, the accurate future data acquisition problem and the uncertainty handling challenges under traditional optimization methods is avoided. At the same time, the spatial-based and time-based clustering methods are applied to improve the accuracy of prediction. Finally, through federated learning, each charging station uploads the local model to the cloud server, and a stochastic client selection pattern is designed to improve the scalability of model aggregation in the cloud server. In this way, the digital assets of each charging station are protected, and computing and communication costs are reduced. Simulation results on real datasets show that the proposed framework has superior performance in terms of training accuracy, communication burden, and computing performance, while maintaining the privacy of EV users and the digital assets of charging stations.

Accurate carbon emission accounting for electric vehicles (EVs) is particularly important, especially for those participating in the carbon market. However, the participation of numerous EVs in vehicle-to-grid (V2G) scheduling complicates the precise accounting of individual EV emissions. This paper presents a novel approach to carbon accounting and benefits distribution for EVs. It includes a low-carbon dispatch model for a distribution system (DS), aimed at reducing total emissions through strategic EV charging scheduling. Further, an improved carbon emission flow accounting model is proposed to calculate the carbon reduction of EVs before and after low-carbon dispatch. It enables real-time carbon flow tracking during EV charging and discharging, then accurately quantifies the carbon reduction amount. Additionally, it employs the Shapley value method to ensure equitable distribution of carbon revenue, balancing low-carbon operation costs and carbon reduction contributions. A case study based on a 31-node campus distribution network demonstrated that effective scheduling of 1296 EVs can significantly reduce system carbon emissions. This method can accurately account for the carbon emissions of EVs under different charging states, and provides a balanced analysis of EV carbon reduction contributions and costs, advocating for fair revenue allocation.