31 May 2023 in IEEE Transactions on Industrial Informatics
False data injection attack (FDIA) can pose a severe threat to the distribution networks (DN), and the accurate detection of FDIA plays a key role in the safe and reliable operation of the DN. In this article, an end-edge-cloud collaboration-based detection framework is proposed to detect FDIA in the DN. First, in order to effectively preserve the privacy of different stakeholders in the DN and solve the problem of data island, a federated-learning-based edge-cloud collaboration mechanism is designed according to the proposed end-edge-cloud collaboration framework to jointly train the local FDIA detection models and eventually build a comprehensive FDIA detection model. Then, considering the temporal–spatial correlation of measurement data, a local data-driven FDIA detection model is proposed based on a novel temporal–spatial graph convolutional network, which can extract temporal–spatial features of the measurement data and improve the FDIA detection performance. In general, compared with the traditional centralized FDIA detection methods, the proposed method can make full use of the computational capacity of distributed edge devices and reduce the pressure of computation on the control center. Finally, simulation results based on the modified IEEE 14-bus and IEEE 118-bus distribution systems indicate that the proposed method can effectively improve the accuracy of FDIA detection compared with other methods.
15 January 2025 in IEEE Transactions on Industrial Informatics
The intermittency of solar irradiance is the main cause of rapid fluctuations in the power output of photovoltaic (PV) systems. These fluctuations hinder the large-scale integration of solar power generation equipment into the grid, which in turn hinders the process of utilizing solar energy resources to reduce carbon emissions. The main way to solve this dilemma is to achieve high-precision forecasting of solar irradiance. Although various methods exist to forecast the variations of solar irradiance, few focus on fully utilizing multimodal data information and fuzzy method to improve the forecasting performance. Therefore, a forecasting method combining multimodal data fusion and fuzzification is proposed to forecast ultra short-term global horizontal irradiance (GHI). First, a modal conversion method is designed to convert temporal modal data to spatial modal data. Then, the fused data are formed by fusing the converted data with normal and under exposure all-sky images. Subsequently, the fuzzy method is used to generate fuzzy GHI data with low nonlinear features. Last, we utilize deep neural networks to learn potential patterns between fused data and fuzzy GHI data in an end-to-end manner. Our method has been comprehensively validated on data provided by the National Renewable Energy Laboratory, demonstrating its effectiveness, and achieving the highest forecasting accuracy compared to state-of-the-art methods.
10 February 2025 in IEEE Transactions on Sustainable Energy
The aggregation and control of massive electric vehicles (EVs) are crucial for grid frequency regulation (FR). However, challenges such as disordered charging, high computational and communication burdens need to be addressed. To this end, a hierarchical hybrid modeling and switching control method for EV aggregation (EVA) is proposed. For modeling, a hybrid state set for EVs comprising three discrete states and one dynamic state is established at the local level. The dynamic state's flexibility allows EVs to charge orderly while considering user demands. At the aggregation level, a Markov-based EVA state space model is designed, integrating the user's willingness-to-pay (WTP) index and hybrid state. It estimates the EVA's FR capacity (FRC) with a lower communication burden and reduces computational burden by simplifying control dimensions. For control, a model predictive control (MPC)-based state switching method is designed at the aggregation level, considering user's FR willingness and power cancellation issue. Furthermore, a predictive compensation mechanism is designed to address model parameter errors resulting from asynchronous control cycles. At the local level, a probabilistic response method is proposed for responding to dispatched control signals, which reduces battery degradation through the state of charge (SOC) based response probability generation. Simulation results validate the method's effectiveness.
7 October 2024 in IEEE Transactions on Transportation Electrification
This paper proposes an end-edge-cloud collaboration based electric vehicles (EVs) aggregator control method that aligns with the EVs-charging stations (CSs)-aggregator architecture of EV aggregator for multiple frequency regulation (MFR), including primary (PFR) and secondary frequency regulation (SFR). This method uses frequency regulation capacity (FRC) as the criterion for calculating MFR tasks, and through the proposed FRC assessment method, the user charging demand is satisfied and battery degradation is reduced during MFR. For PFR, a dynamic droop control based on FRC is proposed at the edges, while for SFR, proportional power sharing is proposed in the edge-cloud collaboration. These methods facilitate efficient utilization of FRC in calculating the MFR task for each CS. However, the disparity in time scales between PFR and SFR introduces a power overrun issue, compromising the accuracy of MFR. To overcome this challenge, a consensus-based edge-edge collaboration mechanism is developed to redistribute the overrun tasks in a distributed way. Furthermore, a two-stage model predictive control is proposed in end-edge collaboration, accounting for the FRC and response characteristic differences between fast and slow charging EVs, to allocate the MFR task from CSs to EVs. Finally, the validity of this method is verified by simulation.
7 August 2025 in IEEE Transactions on Smart Grid
The uncertainty prediction of electric vehicle charging station load plays a crucial role in ensuring the stable operation of power grids. In this paper, a charging station load uncertainty prediction method based on multi-dimensional quantile spatial-temporal traffic flow forecast is proposed. First, considering the complex spatial-temporal correlation characteristics of traffic flow, a traffic flow forecast model based on multi-dimensional spatial-temporal graph convolutional network is designed. The model can deeply explore spatial-temporal correlation characteristics and effectively capture both the static spatial correlations with fixed topological structures and the dynamic spatial correlations of real-time traffic flow. Second, taking into account the importance of prediction results under different confidence levels, this paper combines the forecast model with a weighted quantile regression model and introduces a novel weighted loss function, which can achieve traffic flow uncertainty prediction without the need to predefine the uncertainty probability density distribution and meet the differentiated requirements for uncertainty prediction in grid operation and business decision-making. Furthermore, considering the influence of factors such as road congestion and electricity prices on charging load, a charging station load uncertainty prediction model based on an enhanced queuing model is proposed, which can more accurately convert traffic flow uncertainty into load uncertainty, and thus more accurately simulate the charging station loads. Finally, the simulation results verify the effectiveness of the proposed method.
1 July 2014 in Knowledge-Based Systems