Advanced metering infrastructure (AMI) is the backbone of the next generation smart city and smart grid; it not only provides near real-time two-way communication between the consumers and the energy systems, but also enables the third parties (TPs) to provide relevant value-added services to the consumers to improve user satisfaction. However, the existing services are implemented in a centralized manner, which has potential and associated security and privacy risks also increased with Internet-of-Things (IoT) devices. To better balance the quality of the services and ensure users’ privacy, a TP AMI service model based on differentially private federated learning (FL) is proposed in this article. Instead of sending the private energy data to the cloud server, the proposed service model trains the neural network models locally, and only model parameters are shared with the central server. Moreover, the identity of individuals is eliminated by adding random Gaussian noise during the secure aggregation. Furthermore, an attention-based bidirectional long short-term memory (ATT-BLSTM) neural network model is adopted to solve the long-range dependency problem of conventional neural networks. In the case study, a residential short-term load forecasting (STLF) task is implemented to evaluate the performance of the proposed model. Compared with other state-of-the-art energy service models, the proposed one can achieve similar accuracy as the typical centralized model and balances the trade-off between privacy loss and prediction accuracy flexibly.

A precise short-term load-forecasting model is vital for energy companies to create accurate supply plans to reduce carbon dioxide production, causing our lives to be more environmentally friendly. A variety of high-voltage-level load-forecasting approaches, such as linear regression (LR), autoregressive integrated moving average (ARIMA), and artificial neural network (ANN) models, have been proposed in recent decades. However, unlike load forecasting in high-voltage transmission systems, load forecasting at the distribution network level is more challenging since distribution networks are more variable and nonstationary. Moreover, existing load-forecasting models only consider the features of the time domain, while the demand load is highly correlated to the frequency-domain information. This paper introduces a robust wavelet transform neural network load-forecasting model. The proposed model utilizes both time- and frequency-domain information to improve the model’s prediction accuracy. Firstly, three wavelet transform methods, variational mode decomposition (VMD), empirical mode decomposition (EMD), and empirical wavelet transformation (EWT), were introduced to transform the time-domain demand load data into frequency-domain data. Then, neural network models were trained to predict all components simultaneously. Finally, all the predicted data were aggregated to form the predicted demand load. Three cases were simulated in the case study stage to evaluate the prediction accuracy under different layer numbers, weather information, and neural network types. The simulation results showed that the proposed robust time–frequency load-forecasting model performed better than the traditional time-domain forecasting models based on the comparison of the performance metrics, including the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean squared error (RMSE).

The use of a resistive-type superconducting fault current limiter (R-SFCL) in the nanao modular multilevel converter based multiterminal high-voltage direct current transmission project (MMC–MTdc) is the first of its kind over the world. There were no relevant development and operational experience to support the construction of this project. In response to the difficulties in implementation, the corresponding methods were straightened out in this article. These contributions include 1) proposal of an iterative optimization calculation method for the time-varying quenching resistance of the R-SFCL; 2) the establishment of a simulation platform based on PSCAD/EMTDC for MMC–MTdc system with a nonlinear R-SFCL model; 3) two-group with three-outlet-terminal topology scheme for R-SFCL device; 4) dc-side short-current calculation and short-current suppression rate analyses for MMC–MTdc with the developed R-SFCL; and 5) steady-state operation, insulation performance, and power switching tests of the developed R-SFCL. Conclusively, the feasibility of the R-SFCL prototype is verified with the simulation and engineering practice of pole-to-ground test in the nanao MMC–MTdc system. In conclusion, the presentation of these research contents and conclusions illustrates the future value of industrial experience in the design and operation of R-SFCL MMC–MTdc systems.

The flexible interconnection device (FID) offers significant advantages for interconnecting different distribution networks flexibly. This paper focuses on the significant advantages offered by the FID for interconnecting different distribution networks flexibly. It specifically delves into FID‐based multi‐voltage and multi‐substation distribution networks, proposing a preferable scheme applicable to major load‐centred cities. Beginning with an analysis of constructed FID‐based flexible interconnected distribution network projects, key configurations and features are summarized. Subsequently, typical configurations, electrical parameters, facilities, relevant power functionalities, and application scenarios of multi‐voltage multi‐substation distribution networks are outlined. Building upon this foundation, a suitable interconnection scheme tailored for current urban use is explored to meet the specific needs of load‐centred cities, while incorporating recent advancements in high‐power‐density IGCT technology. An EMT model of a 10 kV/10 MW IGCT‐based four‐substation distribution network is developed in PSCAD/EMTDC. Through thorough analysis under different conditions, the operational performance and benefits are evaluated, providing insights into the efficiency and resilience of the proposed FID‐based interconnection. Lastly, challenges and prospects are discussed from various perspectives to advance the development of FID‐based flexible interconnection solutions. This study aims to contribute to the advancement and implementation of robust interconnection solutions to meet the evolving needs of major load‐centred cities.

With the expansion of smart grids, and the increasing need for efficient infrastructure management, unmanned aerial vehicles (UAVs) have become vital for tasks such as object detection and smart grid inspections. However, current UAV-based systems face challenges related to environmental factors (e.g., fog, wind, and low visibility) and limited computational power, which restrict the use of advanced computer vision techniques. This work presents a lightweight and robust YOLOv8 (LR-YOLOv8) model designed for consumer electronics applications, bridging the gap between advanced computer vision algorithms and real-time operations on resource-constrained UAVs. The model integrates a lightweight architecture, a soft thresholding strategy to handle noisy data, and hue, saturation, and value (HSV) augmentation to simulate complex environments. Additionally, a multiscale domain-adapted loss function improves the mean average precision (mAP) and robustness. The experimental results demonstrate that, compared with the existing object detection methods, LR-YOLOv8 reduces the model weight, and increases the number of frames per second (FPS), while achieving a higher mAP.

Urban Railway Transit Power Supply System (URTPSS) involves power quality problems related to reactive power and harmonics. In addition to this, it is susceptible to multi-system resonance at specific frequencies. Negative impedance caused by capacitive characteristics carries the risk of harmonic amplification, which may lead to equipment damage and instability within the metro substation system. This paper analyzes the monitored profile of the real-world “SHL” metro substation and identifies the capacitive harmonics of the 35kV AC bus. A theoretical analysis is performed and a dynamic model of a Metro Traction Step-Down Substation (MTSDS) is developed to study capacitive harmonics and their potential causes, and to analyze the network impedance across multiple frequencies. Following that, a Distribution Static Synchronous Compensation (DSTATCOM) solution is proposed, featuring a 35kV-10MVar power stage configuration and its associated controller design. The improved instantaneous power controller utilizes the inner α-coordinate current loop to suppress harmonic magnitudes and reconstruct the capacitive impedance. The β-coordinate current is utilized to regulate reactive power demand and correct the system power factor. Simulation results performed in the PSCAD/EMTDC and RSCAD/RTDS environments validate the effectiveness of the DSTATCOM solution with the improved controller under steady-state and load-changing operating conditions.