This paper presents a novel method to optimally place capacitors in power systems that incorporate renewable energy resources and sensitive non-linear loads, considering system uncertainties. A voltage severity index is introduced to assist system operators in the planning and programming of capacitors. The effectiveness of the proposed method is evaluated on a real large copper mine network in Iran, equipped with sensitive loads to voltage sags and doubly fed induction generators (DFIGs). The results indicate that the control mode of DFIGs and their output power influence the optimal location and size of capacitors. The proposed capacitor placement method enhances voltage profiles, reduces total harmonic distortion, minimizes grid losses, and lowers the costs associated with capacitors. Furthermore, the results underscore the necessity of selecting the appropriate control method based on the system’s operational priorities and conditions. The proposed capacitor placement method successfully reduced financial losses due to voltage sags up to 62%, leading to more secure operation of sensitive loads.
Unmanned Aerial Vehicles (UAVs) have gained significant growth and demand across various applications in recent years. Battery-powered UAVs, however, face challenges due to limited endurance. An innovative alternative solution is a fuel cell hybrid power system, but it requires an efficient energy management system (EMS) to coordinate power distribution to meet load among hybrid sources. The drone application’s capabilities are restricted by its limited computational resources; thus, the development of lightweight computation EMS methods is required. This paper proposes a computationally efficient random forest (RF)-based EMS that is trained on data extracted offline from another optimization-based EMS in order to optimize the utilization of hybrid sources and to coordinate the power distribution at high computational efficiency. Simulation demonstrated the strategy’s efficacy; it reduces the battery state of charge (SOC) deviation, which extends the battery durability, and it also maintains a stable DC bus voltage stability close to its reference value. The strategy successfully reduces hydrogen consumption, showing economical and technical benefits at low computational demand.
Accurate real-time control of three-phase inverters in AC microgrids is challenged by the need to effectively manage multi-variable control while meeting tight sampling instant time limit. Model predictive control (MPC) offers efficient constraint handling, but it often suffers from computational complexity. To address this, this paper implements a random forest (RF)-based controller that is trained offline with extensive historical data extracted from MPC. Thereby reducing computational demands and improving adaptability to varying loading conditions. The proposed RF-based controller is assessed under various operating scenarios of the AC microgrid, including resistive, inductive, and capacitive loading conditions, demonstrating its ability to maintain high-quality sinusoidal output voltage with low total harmonic distortion (THD). The findings prove that this approach not only preserves the advantages of MPC but also significantly enhances computational efficiency, offering a promising solution for modern power electronics applications.
To improve the flight reliability of distributed electric propulsion unmanned aerial vehicle (DEP-UAV), a novel fault-tolerant control strategy for T-type three-level inverters is proposed in this paper, and the weighting factor tuning of finite control set model predictive control (FCS-MPC) is implemented based on long short-term memory network (LSTM). Compared with a conventional fault-tolerant control strategy, the proposed method could effectively solve simultaneous failures of multiple switching devices and quickly adjust the weighting factors under multiple constraints changes to ensure stable system operation. The method’s feasibility has been verified by simulating the operation of DEP-UAV in a full numerical simulation platform.
In many applications of wireless power transfer (WPT), it would be desirable to activate the primary or transmitter side power stage only when the receiver or secondary side is present. This feature is usually achieved by employing some form of active communication or receiver-side modulation to inform the transmitter side controller of the presence of receiver side, which increases the complexity of the WPT system and raises concerns over reliability in noisy environments. In this paper, a communication-less receiver detection scheme based on S-LCC topology is proposed. The proposed receiver detection method for WPT systems is not affected by load condition, and provides a near constant output voltage at receiver side; except for output short-circuit condition, in which case the transmitter side would not be activated in order to protect the system. The proposed system and method are verified experimentally with a hardware prototype that successfully demonstrates receiver detection and short-circuit protection functionality.
In this paper, a novel coordinated control is proposed to achieve integrated generation of the solid oxide fuel cell-gas turbine (SOFC-GT) system with consideration of the ambient temperature effect (ATE), which equips the system with both grid following (GFL) and grid forming (GFM) capabilities under safe and efficient operation. Moreover, how the integrated generation would be influenced by the ATE has also been analyzed. Key control algorithms and validations based on a controller hardware-in-the-loop (C-HIL) testbed are provided.
This study investigates the optimal planning and operation of distributed energy resources (DERs) with energy management scheme (EMS) in a microgrid (MG). The optimal allocation of DER units and scheduling of the shiftable loads are carried out to improve the overall investment cost of generation, power loss, and reliability of the MG. The expected outage cost (ECOST) of customers is evaluated to measure the reliability of the MG. A multi-objective particle swarm optimization (MOPSO) technique is used to solve the proposed multi-objective planning problem. To select the best compromise solution from the Paretooptimal front, a fuzzy optimization approach is used. Several case studies are conducted and compared to validate the proposed long-term planning methodology. Simulation results prove the efficacy of the proposed model.
This study aims to enhance the efficiency of underwater wireless power transfer (WPT) stations, enabling Autonomous Underwater Vehicles (AUVs) to recharge efficiently while reducing reliance on surface supply stations. To achieve this goal, COMSOL Multiphysics® will be integrated with MATLAB to establish an evaluation platform for analyzing the contribution of ferrite plates to the magnetic field. The coil and ferrite plate models constructed in COMSOL Multiphysics® are imported into MATLAB, where parameter adjustments can be flexibly applied and mutual inductance is extracted for further evaluation. Using the Random Forest (RF) algorithm, ferrite plate positions with the highest contributions are identified and retained. The optimized ferrite plate placement is applied in a WPT system, achieving a 4.4% improvement in efficiency compared to the design without ferrite plates, and reducing the total weight by nearly 33% compared to the design with fully-covered ferrite plates.
This paper proposes a non-cascaded fractional order sliding mode (FOSM) control strategy for the brushless doubly-fed induction generator (BDFIG) to compensate the mismatched uncertainties structured by parameter perturbations when the temperature and frequency of the system change. The mismatched uncertainties in the mathematic model of BDFIGs are analyzed and modeled. For the second order nonlinear system with mismatched uncertainties, a fractional order sliding surface and chattering-free control law are designed in the non-cascaded controller. The accuracy, dynamic response and robustness against the uncertainties in the BDFIG control system have been enhanced. The effectiveness and superiority of the proposed control strategy are proved by experiments.
As the share of renewable energy in power systems increases, the equivalent system inertia will gradually decline, making it more difficult to maintain stable frequency during sudden changes in load or generation. This can result in challenges such as higher Rate of Change of Frequency (RoCoF), and more significant fluctuations during extreme frequency events. To address these issues, future power systems must treat the rotating mass within the grid as a critical component, ensuring a minimum level of inertia is always maintained. Additionally, alternative inertia compensators should be considered as backup solutions. This survey explores key issues related to system inertia response, inertia estimation, and frequency stability, with a focus on grid code amendments in regions with high renewable energy penetration. Specifically, it addresses grid codes concerning RoCoF characteristics and revisions to operational standards. This work provides valuable insights into future grid code requirements and development of auxiliary service market, supporting system operators in ensuring the safe and reliable operation of grids with substantial renewable energy integration.
Accurate modelling, parameter extraction, and State of Charge (SoC) estimation of Li-ion batteries are essential for grid-connected Electric Vehicle (EV) applications to ensure optimal energy management, extended battery lifespan, and grid stability. This paper presents a two-stage approach involving simulation-based SoC estimation and experimental parameter extraction for Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) operations. Initially, a MATLAB/Simulink model is developed featuring a photovoltaic system, Battery Energy Storage System (BESS), and a grid-connected EV battery. SoC estimation is conducted using Coulomb Counting, Extended Kalman Filter (EKF), and Unscented Kalman Filter (UKF). Separately, a pulse discharge test is performed using MATLAB and dSPACE to extract key internal parameters such as ohmic resistance, polarization resistance, capacitance, and open-circuit voltage. These parameters are then utilized in a MATLAB-based framework to estimate SoC under simulated load profiles. This study bridges simulation and experimental insights, contributing to more accurate and reliable battery management strategies for renewable-integrated EV systems.
Accurate solar energy forecasting is crucial for grid stability and renewable energy integration, but missing data significantly impacts forecasting accuracy. Traditional statistical imputation techniques often fail to capture the complex, nonlinear nature of solar energy variations, necessitating advanced machine learning approaches. This study addresses this critical challenge by proposing a Multi-Input Fuzzy Rules Emulated Network (MiFREN) model for efficient and accurate reconstruction of missing solar power data. Beyond mere imputation, MiFREN leverages expert knowledge through its interpretable fuzzy rule structure, enabling more robust and physically meaningful forecasts. MiFREN demonstrably outperforms benchmark AI models (KNN, Random Forest, LSTM), achieving the lowest prediction error (NMAPE 1.17 percent, NRMSE 4.13 percent). Crucially, this superior performance is achieved with remarkable efficiency: MiFREN's structured rule base significantly reduces model complexity, requiring only 27 weight parameters – a fraction of the parameters needed by comparable models. This inherent efficiency makes MiFREN highly practical for real-time applications. These findings highlight MiFREN's unique capabilities for enhancing solar energy integration, improving grid reliability, and providing valuable insights into solar power dynamics through its transparent and readily understandable rule base.
In smart grids, differential current relays protect power transformers by comparing local and remote measurements communicated over substation networks. However, this reliance on communication makes them vulnerable to false data injection attacks (FDIAs), leading to false tripping of the protected transformer and possibly system instability. This paper proposes a novel, quantum-based, data-driven scheme for detecting FDIAs targeting transformer relays. The proposed approach utilizes quantum variational circuits (QVCs) to analyze relay measurements, accurately distinguishing between malicious measurements and those associated with genuine fault conditions. The proposed scheme is trained and tested under various FDIA and fault scenarios generated in an OPAL-RT environment. Our results demonstrate that the proposed QVC-based scheme accurately detects FDIAs, maintains relay dependability, and outperforms existing solutions. The proposed QVC scheme is also validated using an OPAL-RT Hardware-In-the-Loop real-time simulation.
This paper presents the design, analysis, and experimental validation of a high-efficiency single-stage primary-side feedback flyback converter specifically optimized for LED lighting applications. The converter leverages primary-side regulation techniques to achieve constant output voltage control without the need for secondary-side feedback components, such as optocouplers, effectively minimizing circuit complexity, size, and manufacturing costs. The proposed approach integrates variable switching frequency and adaptive valley switching methods to significantly reduce switching losses and electromagnetic interference (EMI). Operating in boundary conduction mode (BCM) and discontinuous conduction mode (DCM), the converter demonstrates robust performance under wide input voltage variations from 90V to 277V AC, maintaining high efficiency and exceptional power factor correction (PFC). Extensive simulations performed using SIMetrix/SIMPLIS validate the theoretical analysis, which is further supported by practical measurements. Experimental results indicate peak efficiency exceeding 92%, with a maximum measured power factor of approximately 0.997 under full-load conditions. This research not only addresses critical challenges in reducing energy consumption and environmental impact associated with electronic waste but also offers a detailed step-by-step design methodology for real-world deployment. Consequently, the findings provide significant insights and practical guidance for developing advanced, compact, and reliable flyback converters suitable for next-generation energy-efficient lighting solutions.
Recent research on GFMCs (Grid-Forming Converters) with high-penetration RESs (Renewable Energy Sources) highlights their ability to mimic SGs (synchronous generators), enabling RESs to self-synchronize, share power, and support grid frequency response. However, many GFM control methods and loops significantly alter short-circuit fault characteristics, affecting traditional protection systems, especially for the functions of distance protection relays in transmission grids. Distance relays operate by calculating impedance from local voltage and current. GFMCs, with current limiters and FRT (Fault Ride Through) controllers, will modify the amplitude and phase angle of reference currents, reducing impedance below relay settings. This change affects relay sensitivity and reliability in detecting short-circuit faults. Thus, this article aims to explore the impact of GFMCs on relay protection in transmission grids and provide insights for designing new protection systems that can adapt to evolving power grids, where GFMCs and GFLCs (Grid-Following Converters) play a key role.
Extensive research has focused on the control strategy and capacity ratio of the converter side to prevent resonance problems, ignoring the line overload issues and the regulation of the grid side on the resonance stability. To this end, a source-grid coordinated planning model embedded with resonance stability constraints is developed in this paper. First, a linearized resonance stability constraint model is established based on an impedance analysis method and the positive net damping criterion. Then, a grid topology adjustment model is constructed based on the circuit equivalence method, and a source-grid coordinated planning model that can satisfy the resonance stability constraints is proposed. The validation results of the IEEE-24 case system show that the resonance stability capability of the system can be improved by the reasonable grid topology adjustment, and the source-grid planning scheme can optimize source allocation based on changes in the grid topology.
Non-intrusive load monitoring (NILM) is an advanced technology for intelligent energy management. Although graph signal processing (GSP) concepts have been applied to NILM in an unsupervised way, the performance of such solutions remains unstable and undesirable. In this paper, a new unsupervised NILM framework is proposed. The original state transition sequence (STS) extraction method is first improved. Then, the operational duration is introduced as a novel time-wise feature. This feature is fused with power-wise features and utilized in both pairing and clustering processes. Results on open-access residential and industrial datasets indicate that the proposed method significantly outperforms other benchmarks, making it promising for practical implementation.
This paper presents an adaptive Kalman filtering approach for real-time detection and characterization of power oscillations in grids with mixed synchronous and inverter-based generation. The methodology combines dynamic autoregressive modelling with recursive parameter estimation to track damping ratios and frequencies of inter-area oscillations (0.1-2 Hz). Validated against System Protection and Dynamics WAMtool software and tested under diverse scenarios including simulated environment (Kundur two-area system) and real network data, the algorithm achieves high accuracy in damping estimation. The work bridges model-based and measurement-based approaches, enabling real-time stability assessment without the need to rely on full system models.
Comprehensive evaluation of motor cost, performance, and thermal tradeoffs is crucial for rare-earth permanent magnet (REPM) reduction design in permanent magnet synchronous machines (PMSMs). This paper investigates the impact of proportional REPM reduction on torque and armature current characteristics through a case study of a double-layer V-type IPMSM. A functional relationship between torque and parameters including current, inductance, and flux linkage is derived based on the maximum torque per ampere (MTPA) control strategy. By treating motor inductance and flux linkage as constants, quantitative correlations between torque, current, and the flux weakening ratio (FWR) are established. The mapping between REPM reduction ratios and armature current requirements is analyzed across full speed ranges. Furthermore, variations in power factor and efficiency under different REPM reduction levels are systematically examined. A three-port three-dimensional lumped parameter thermal network (TPTD-LPTN) model considering non-uniform temperature distribution in rectangular conductors is developed to evaluate stator winding temperature variations under multiple operating conditions with varying REPM content. The inverse correlation between REPM reduction ratios and winding hotspot temperatures is quantified, while the spatial distribution and migration patterns of thermal hotspots are revealed. These findings establish quantitative relationships between REPM reduction proportions and winding temperature rise penalties, providing theoretical guidance for low-REPM PMSM designs.
A hybrid Convolutional Neural Network (CNN)-Vision Transformer (ViT) method is introduced to identify early stator turn-to-turn short-circuit (TTSC) faults in induction motors (IMs) driven by variable frequency drives (VFDs). By combining global self-attention with local spatial feature extraction, it enables simultaneous severity evaluation and faulty phase determination. To eliminate VFD-induced harmonics and noise, current signals undergo low-pass filtering and Z-score normalization for standardization. Next, time-series data is transformed into high-resolution two-dimensional images using the combination of Local Weighted Stockwell Transform and Adaptive Chirplet Transform in order to improve feature extraction. Subsequently, the hybrid CNN-ViT model is trained and validated utilizing the generated images using 10-fold cross-validation to ensure generalization and minimize overfitting across all cases. Data set utilized for training validation derived from the PSCAD simulation generated over a variety of TTSC fault severities, loads, and fault resistances. The suggested approach enhances diagnostic reliability and supports predictive maintenance, reducing unplanned industrial downtime.
