In this study, a framework combining multiple deep learning techniques is proposed for electrical equipment state recognition and monitoring to address the problem that existing methods have limited feature extraction capabilities under high noise and complex working conditions. The adaptability of the model is improved through data augmentation and self-supervised contrastive learning. A hybrid architecture of CNN-BiLSTM and Transformer is designed to extract spatiotemporal features, and the model performance is optimized by combining domain adaptation technology, neural architecture search (NAS), and deformable convolutional network (DCN). The experimental data comes from a large-scale electrical equipment monitoring system in an industrial park in a certain province, covering 15 equipment states and a total of 269,000 multimodal data. The experimental results show that the proposed method is significantly superior to the baseline model in terms of recognition accuracy (95.37%), real-time performance (detection delay of 3.02ms), and cross-domain adaptability (improved by 41.5%), providing an efficient and reliable solution for electrical equipment state monitoring, which has important theoretical and practical application value.
Driven by the global energy crisis and carbon neutrality goals, Building-Integrated Photovoltaic (BIPV) systems are increasingly recognized as a key solution in sustainable architecture. This study presents a comprehensive review of BIPV integration approaches and emerging performance strategies at the building scale. Combining qualitative analysis with bibliometric methods, the research is based on publication data from the Web of Science Core Collection, using relevant topics in photovoltaic technology and green building, and covering the years 2001 to 2025. The paper traces the evolution of BIPV technologies, identifies research hotspots, and reveals shifts in structural and performance paradigms. Technically, BIPV systems have evolved from simple energy-generating add-ons to multifunctional building envelope components that integrate structural, protective, visual, and environmental regulation functions, with applications in roofs, façades, shading systems, and more. Performance assessment has become increasingly multi-dimensional, encompassing energy conversion efficiency, thermal management, environmental impact, aesthetic integration, and economic viability. Bibliometric analysis reveals a transition from structural integration to multi-objective performance optimization, with smart modeling, AI-assisted design, and urban energy network integration emerging as future directions. Despite this progress, BIPV implementation still faces systemic barriers such as structural complexity, high costs, regulatory fragmentation, and limited interdisciplinary collaboration. In response, this paper proposes a framework addressing intelligent modeling, policy incentives, standard development, and design education to advance the deep integration and widespread adoption of BIPV in green buildings. The findings aim to provide theoretical insights and practical guidance for BIPV deployment, policymaking, and future research.
In view of the problem that the current power grid dispatching voice interaction system is not adaptable enough to the terminology specific to the power grid dispatching field and is easily disturbed by environmental noise, resulting in command recognition errors and missing keywords, this paper constructs a deep fusion model based on Wav2Vec 2.0 self-supervised pre-training and Conformer structure, aiming to achieve Automatic Speech Recognition (ASR) and optimize accuracy. First, based on the Wav2Vec 2.0 model, the original dispatching voice signal is self-supervised pre-trained to extract features and capture its low-level time domain and frequency domain expressions. Then, the extracted voice features are input into the Conformer structure fine-tuned by the dispatching field corpus to achieve high-precision modeling of long-distance context. Finally, the power grid professional terminology dictionary is embedded in the decoding stage, and the spectrogram enhancement and background noise synthesis mechanism are combined to achieve end-to-end joint optimization. The results showed that the accuracy, recall, and F1 score of the speech recognition model in this article were 92.3%, 89.1%, and 90.7%, respectively, with an average of Word Error Rate (WER), Character Error Rate (CER), Weighted WER were 10.8%, 5.7%, and 13.8%, respectively; The F1 score for term recognition reached 90.7%; The recognition rate of Top-3 is above 0.75, and the complete recognition rate of instructions reaches 84.6%. Under extreme low signal-to-noise ratio conditions of -5dB, its WER control is 42.1%. The conclusion shows that the method proposed in this paper can effectively improve the accuracy and scene adaptability of ASR, provide reliable support for high-precision voice interaction in power grid scheduling, help improve the safety and reliability of power facility operations, and reduce work delays caused by misoperation or poor communication.
The use of different types of renewable energy sources has been growing very rapidly recently. This is especially true for the use of wind and solar energy to produce heat and electricity. Many advanced countries have already achieved a share of renewable energy sources in the total energy production balance of about 40-50 per cent. This is quite high, although the goal of reducing the level of CO2 in the earth's atmosphere has not yet been achieved. The unstable over time capacity of generation using renewable energy sources leads to additional difficulties in ensuring the reliability of power supply and the quality of generated electricity, which eventually leads to the need for widespread use of energy storage technologies. The most widespread is the system of hydro-pump energy storage. It is practically the only technology to date that allows storing large amounts of energy for a long period of time. The purpose of this study is to analyse in depth the operating experience of hydro pump energy storage, quantify the efficiency of their operation, and develop the most promising directions to improve their flexibility and efficiency. A detailed efficiency analysis is performed on the example of the hydro pumped storage power plant "Gorona del Viento" (El Hierro Island, Canary Archipelago, Spain). Possible methods of load balancing in the electric network by means of a hydro pumping power plant are considered. Using computational models, the main operating modes of the power plant are analysed, and the possibility of implementing different strategies of load balancing in the electric grid is demonstrated. Key words. Hydro pumped energy storage, efficiency, power plant flexibility, renewable energy.
This paper presents an application of deep reinforcement learning (DRL) for controlling permanent magnet-assisted synchronous reluctance machines (PMA SynRMs). A model-free DRL agent is trained to control the power converter switching states, aiming to accurately track current references. The DRL-based control scheme is compared against a traditional finite control set model predictive control (FCS-MPC) strategy employing a simplified linear model of the PMA-SynRM. Simulation results demonstrate that the DRL controller achieves superior performance in terms of tracking accuracy and harmonic distortion reduction, effectively handling the machine's inherent nonlinearities. Furthermore, the DRL agent exhibits robustness against measurement errors. The findings highlight the potential of DRL as a viable alternative to conventional model-based control methods for high performance PMA-SynRM drives, offering improved adaptability, robustness, and operational flexibility. Key words. Artificial neural networks, deep reinforcement learning, electrical drives, model predictive control, permanent magnet-assisted synchronous reluctance machines.
Enhancing the energy efficiency of buildings, roads, or other applications involves, as a main challenge, the development of innovative materials with customised thermomechanical properties and thermal energy capabilities while maintaining cost effectiveness. These materials can be designed to find application as solar heat absorbers or to prevent issues related to urban heat island. To that end, a reliable assessment of the material behaviour under solar irradiation is required. With this aim, this work proposed an experimental setup that has been simulated by CFD and built to test different bituminous and non-bituminous binders with potential use in roofing, road and heat collectors. Key words. Bitumen, non-bituminous binder, solar behaviour, CFD.
The short measurement periods of local solar resource measurement campaigns limit the representativeness of these measurements in long-term energy production estimates for solar projects. This study aimed to characterize the main long term solar resource databases and correlate them with data provided by meteorological stations, with the purpose of identifying the database that best predicts solar radiation at specific locations. Seven long-term databases were used (PVGIS, SOLARGIS, SOLCAST, NASA POWER, NSRDB, HELIOCLIM, ERA5). Measured data from meteorological stations at various locations in Portugal, Saudi Arabia, and Brazil were correlated with the long-term databases available for each location. Different methodologies for evaluating these correlations were tested, including R², MSE, MBE, and MAPE. The analyses revealed that the SOLCAST and SOLARGIS databases highlighted in predicting global horizontal irradiance (GHI) in the Arabia region. In Portugal, the NSRDB was the most accurate in predicting GHI, while in Brazil, SOLCAST showed the highest accuracy in forecasting GHI. Key words. Solar resource, local measurements, long term databases, error metrics
Public lighting plays a major role in the global figures of energy consumption. Since its final goal is to ensure the safety of people and goods as well as the well-being of its users, it is a clearly basic service. Although the continuous progress in more efficient and sustainable light sources seem to leave a large margin of improvement in terms of energy efficiency, its safety-related peculiarities of public lighting, especially in infrastructures like streets, roads and tunnels, make it a challenge to get good visual performance with minimal energy consumption. Regulations and standards have proposed different coefficients to take account of this efficiency in public lighting with discrete success yet. This work discusses the current situation and presents some proposals with the target of making more sustainable cities without compromising the safety of their citizens.
Sub-synchronous oscillations pose a significant challenge in modern power systems, particularly in networks with high penetration of power electronic converters. While phasor measurement units (PMUs) are generally used for grid monitoring, their ability to detect sub-synchronous oscillations is being challenged. This paper discusses PMU limitations in measuring sub-synchronous oscillations, introducing a novel algorithm to detect their occurrence by leveraging PMU phasor data and rate of-change-of-frequency analysis to trigger high-resolution voltage waveform recordings. This approach enables direct comparison between PMU-reported events and raw waveform data, which paves the way to analysing discrepancies and limitations in PMU based detection. Key words. Measurement techniques, power grids, phasor measurement units, oscillations, power quality.
A rough set combined with an ant colony algorithm is used to extract the non-forced vibration signal from the measured transmission line galloping displacement signal, which is used as the risk feature recognition index. In the real model test, the transmission line's six split line segment shift time history is measured across multiple channels to identify the galloping frequency, vibration mode, and damping ratio of the transmission line. At the same time, MATLAB simulation is employed to verify the algorithm's accuracy and computation time. The results show that the accuracy of the proposed algorithm in identifying the galloping frequency, vibration mode, and damping ratio is 98%, 97.6%, and 98.5%, respectively, with calculation times of 16s, 18s, and 17.5s. Therefore, in multi-channel modal analysis, a rough set combined with the ant colony algorithm can effectively address the problem of dangerous feature identification of transmission line galloping.
Hardware-in-the-loop (HIL) testing based on real time simulators is one of the critical methods for the power system and the power electronics system research. Using multiple interconnected simulators to enhance parallel computing performance is a common method. However, with the increasing demand for enlarging system scale and shortening simulation time-step, more dedicated communication resources and a higher amount of transferred data are required. This paper proposed a method for optimizing inter real-time simulator communication utilizing the delay variety of the distributed transmission line model, which can reduce the communication hardware resources and increase the data throughput with no loss of the simulation accuracy. In the case study of a system containing 100 wind turbines and 20 sets of IEEE 39-bus system on four real-time simulators, the communication resource consumption is reduced by 48.78% with the proposed optimization method. Key words. Real-time Simulator, Transmission Line Model, Communication Optimization
The CEL RURAL project has been conceived with the objective of promoting the implementation of renewable energy solutions in rural areas, with a view to enhancing the development of sustainable energy systems. The project has three overarching aims: first, to create energy self sufficient Local Energy Communities (LECs); secondly, to optimise energy efficiency; and thirdly, to monitor energy systems for improved operation. In addition, the project focuses on community engagement, fostering innovation, and ensuring compliance with local and international regulations. The project is currently testing microgrid solutions and energy storage technologies, including the use of lithium-ion (Li-ion) batteries in one of the pilot plants, specifically LiFePO4. This is being done through four pilot plants across Portugal and Spain. Keywords. Energy communities, local energy communities (LECs), energy storage, microgrids.
Wind energy has experienced significant growth in recent years; however, it still faces challenges in operation and maintenance, which impact energy efficiency and lead to high costs. This study proposes an anomaly detection model for wind turbines based on a support vector machine (SVM), optimized using Bayesian search. The model was trained using vibration data from the Gearbox Reliability Collaborative (GRC) project of the National Renewable Energy Laboratory (NREL), specifically from the High-Speed Shaft Upwind Bearing Radial and High Speed Shaft Downwind Bearing Radial sensors, with 40,000 records evenly distributed between normal and anomalous conditions. The proposed model achieved an overall fault detection accuracy of 78.95% and 78.50% for the respective sensor data. Bayesian optimization facilitated the fine-tuning of the hyperparameters in the classification technique, enhancing the model's anomaly detection capability. Furthermore, the use of vibration data enabled the identification of critical fault patterns in turbine operation, contributing to the improvement of wind turbine efficiency and reliability. Key words. Support Vector Machines, Anomalies, Wind Turbines, Bayesian algorithm.
To effectively deal with the urgent climate crisis,clean resources and energy efficiency are becoming essential. As a result, renewable energy technologies (e.g., solar PV and wind power), HVDC links, and multi-energy systems have been widely integrated into traditional electrical power systems in recent decades. These technologies and HVDC links provide numerous benefits in driving the energy transition, and multi-energy systems are a promising energy storage/management solution to handle the variable and unpredictable nature of renewable energy technologies. However, despite the many benefits of the above new sustainable energy infrastructures, they also pose new technological challenges, such as small-signal stability problems due to poorly damped resonances caused by the interaction between power electronics and traditional grid elements. Current tools are not fully effective in studying these issues yet. Given this scenario, this paper presents an Innovative Software for Stability Analysis, a novel tool designed for small-signal stability assessment in multi-energy grids. This software enables accurate stability predictions and provides actionable solutions to mitigate instabilities, regardless of system size. Its capabilities are demonstrated through comprehensive MATLAB/Simulink simulations on various systems, including a IEEE 3−bus test system, an AC/DC multi-energy grid with hydrogen vector, and a 70k−bus synthetic test system. Key words: Damping solutions, multi-energy grids, positive-mode-damping, resonance mode analysis, stability assessment.
This paper presents an analysis of the Voltage Source Converter-High Voltage Direct Current (VSC-HVDC) technology in power systems since it was implemented experimentally in the late nineties to the present. The main features as well as the main applications are explained attending to current projects in operation. Additionally, a recopilation work has been made including two Annex. In the first one, all projects commissioned around the world have been summarised. The second Annex comprises multi-terminal projects in operation and planned during this decade. Key words: VSC-HVDC, interconnectors, multi-terminal, back-to-back, hub, offshore wind farm.
Accurate fault detection in wind turbines is essential for maximizing operational efficiency and reducing maintenance expenditures. This paper presents TransWind, a novel Vision Transformer (ViT)-based framework designed specifically for analyzing SCADA data to pinpoint and diagnose issues in wind turbines. Unlike traditional machine learning models, TransWind leverages the attention mechanisms of ViTs to capture complex temporal and spatial relationships within SCADA time-series data, including parameters for example rotational speed, generator temperature, and electricity output. This unique capability allows for precise identification of anomalies and their underlying causes. The innovation of TransWind lies in its capacity to integrate explainable AI (XAI) techniques, including Self-Attention Attribution, to provide transparency in fault predictions, enabling maintenance teams to focus on critical system elements. The proposed framework will assessed on publicly available SCADA datasets, demonstrating a fault detection accuracy improvement of 8-12% compared to state-of-the-art models. Furthermore, TransWind exhibits robustness in handling noisy and incomplete SCADA data, a common challenge in real-world deployments. This research highlights the transformative potential of transformer based architectures in renewable energy fault diagnostics. By enhancing detection accuracy and interpretability, TransWind offers a scalable solution for predictive maintenance, reducing turbine downtime and operational costs while advancing AI-driven sustainability in wind energy systems. Key words. Vision Transformers (ViTs), SCADA Data Analysis, Wind Turbine Fault Diagnosis, Explainable Artificial Intelligence, Predictive Maintenance
The integration of renewable energy sources in power systems has become a stability challenge. Their connection is made via power electronics, mostly using a grid-following control scheme. However, it has been proven that this type of control raises instability problems when they are connected to weak grids. This paper focuses on analyzing the stability of grid following converters when connected to grids with different stiffness using frequency analysis. Firstly, an assessment of the converter’s passivity indicates the frequency range where the grid stability can be compromised. Then, a time-domain simulation shows how the converter responds to a voltage dip in each case. Finally, the frequency-domain impedance on the small-signal model reconfirms the previously obtained results. The conclusions support the theory explained and point possible future applications. Key words. Grid-following converter, weak grid, stability, passivity, impedance stability analysis.
This work aims to develop an efficient technology for valorizing by-products from agricultural and forestry activities by creating high-performance thermal insulators. Complex biodispersions were designed by incorporating phase change materials (PCM) into biofoams for thermal storage and regulation applications. The developed composites demonstrated effective PCM encapsulation, with latent heat storage capacities reaching up to 96.58 J/g for the highest PCM concentration studied. These materials will enhance industrial efficiency in areas such as high-temperature heat recovery, cold energy storage, and building insulation. Additionally, this material not only seeks technological advancement in insulation, crucial for conserving energy by minimizing thermal transfer and improving the safety and quality of industrial processes, but it also stands out as a sustainable technology. The use of bio-based polymers also maximises biomass resources and generating socio economic benefits through employment in agriculture and forestry. Key words. Thermal insulation, bio dispersions, phase change materials (PCM), thermal energy storage, sustainable technology.
The current material warehousing and outbound management of the power system has problems such as low efficiency, rigid planning, and frequent path conflicts, which lead to material accumulation, regional congestion, and reduced operating efficiency, seriously restricting the stable operation of the power supply chain. In response to this situation, this paper proposes a power material storage and scheduling optimization method based on the deep deterministic policy gradient (DDPG) algorithm and the 6G-enabled cyber-physical system (CPS). In the warehousing stage of power materials, combined with the high bandwidth and low latency characteristics of the 6G network, the CPS system realizes real-time perception and dynamic storage adjustment of material information; in the outbound stage, the DDPG algorithm is used to construct a continuous state-action space, optimize the path planning and dynamic obstacle avoidance of automatic guided vehicles (AGV) equipment, and improve the scheduling efficiency of power emergency materials. Experimental results show that after integrating the 6G-CPS system, the material scanning time is shortened from the traditional 45 seconds/item to 25 seconds/item, and the efficiency is improved by 44%; the optimized path distances of DDPG in simple and complex environments are 7.2m and 8.6m respectively, which are better than the comparison algorithms such as A2C (Advantage Actor-Critic) and A3C (Asynchronous Advantage Actor-Critic), and the convergence speed and reward value are optimal. The research results provide an efficient solution for smart grid material management, which can effectively support the material dispatching needs in scenarios such as new energy access and power emergency repair.
Electric Vehicles (EVs) operating on 800V architecture, and their associated fast-charging stations, demand higher power capacity, as compared to traditional 400V EVs. This increases stress on the utility grid. Bifacial photovoltaics (bPVs) offer a high performance energy generation solution, capable of mitigating these demands. They are known for their bidirectional irradiance capture technology, and dual-surface PV conversion capability. Using superior photon-to-electron conversion efficiencies and enhanced power density, bPVs show an advantage over monofacial photovoltaics (mPVs). This paper evaluates the energy generation potential of bPVs, integrated with grid-supplied power, to support 800V EV charging requirements. This is achieved through computational modeling and simulation with PVSyst, using meteorological datasets, from selected regions in Spain. For space constrained urban areas, this paper technically validates that energy output of bPVs represent a superior strategy for supporting the electrical grid, in the charging of 800V EVs, as compared to traditional mPVs. Key words. Bifacial, Albedo, DC Fast Charging, 800V, Electric Vehicle
