Addressing a restoration challenge within a sustainable radial electrical distribution system involves resolving issues to re-energize loads downstream of sectors impacted by a permanent fault. This research article presents a Non-dominated sorting Binary Particle Swarm Optimization method for solving a sustainable electrical radial distribution system's multi-objective in-service restoration problems. The proposed hybridized methodology is faster in computation than other methods, such as the conventional genetic algorithm, non-dominated sorting genetic algorithm, reactive tabu search, and conventional particle swarm optimization. Single- and multiple-fault case studies have been taken to analyse full-service restoration. The simulation study and comparative analysis have been performed on standard IEEE 10-bus (System I), 13-bus (System II), 33-bus (System III), and a practical 173-bus (System IV). The system I, NSBPSO, with and without PFC for the single-fault case, shows a difference in execution time of 25.89 and 4.27 s, respectively. Similarly, for the multi-fault case, the difference in execution time is 10.3 and 5.36 s, respectively. The power losses found from load flow results after applying M1 are 104.50 kW, while M5 is 106.86 kW. The proposed method, M1, executed in less time compared to other methods to proof a reliable and promising solutions in electrical distribution system.
Due to the growing number of electric vehicles (EVs), public charging stations are faced with the need for intelligent scheduling frameworks. The current methods of slot allocation cause high wait times, high power losses and poor user experience. In this manuscript, a hybrid framework is proposed to schedule EV charging sessions in an optimal manner while maintaining a quality-of-service (QoS) guarantee. The uncertainty-aware predictions of the charging session duration for each EV arrival is generated by the Bayesian constitutive artificial neural network (BCANN) component, which allows the scheduler to reserve a time slot for high-uncertainty charging sessions. The Kirchhoff’s law algorithm (KLA) subsequently optimizes charging schedules, using an objective that combines total charging cost with the loss cost. The proposed BCANN-KLA framework is simulated and compared with three benchmark methods such as many-objective stochastic competition optimization algorithm, Chaotic Harris Hawks Optimization algorithm and the artificial neural network based scheduler. The results of statistical evaluation over 30 independent runs have shown that the scheduling efficiency of BCANN-KLA is 98.74
The deterministic day-ahead (DA) market does not accommodate the uncertainty from variable renewable energy sources (VRES), often leading to increased balancing costs. Alternatives based on stochastic programming (SP) and adaptive robust optimisation (ARO) have been proposed in the literature but have yet to be implemented in functioning DA markets. This paper compares the deterministic, SP, and ARO market-clearing approaches for the co-optimization of energy and reserves, highlighting their advantages and drawbacks. Three case studies inspired by the power systems of the Netherlands, France, and Germany are examined using in-sample and out-of-sample analyses. Sensitivity analyses explore the impacts of sample size and conservativeness parameters on model performance. The findings reveal that SP and ARO models enhance socio-economic welfare (SEW) and significantly reduce the need for balancing services after the DA compared to the deterministic approach. Notably, the deterministic model yields the lowest SEW on days with high forecasted VRES output, whereas uncertainty-based models demonstrate the greatest benefits under high VRES penetration. Paving the way towards real implementation, we show that the trade-offs between economic performance, integration of VRES, and balancing needs after the DA depend on both system characteristics and the uncertainty model applied.
Grid-connected photovoltaic (PV) systems require fast and robust maximum power point tracking (MPPT) to maintain high energy yield and grid power quality under rapidly varying irradiance, temperature fluctuations, and partial shading conditions. Conventional MPPT methods such as Perturb and Observe (P O), Incremental Conductance, and shallow neural networks often suffer from slow convergence, steady-state oscillations, and mis-tracking under dynamic operating conditions. To address these limitations, this paper proposes a hybrid deep-learning-based MPPT strategy combining Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and Transformer attention (C-BiLSTM-TrA) for grid-connected PV systems. The CNN extracts spatial correlations among PV voltage, current, irradiance, and temperature, the BiLSTM captures bidirectional temporal dependencies, and the Transformer attention mechanism models long-range contextual relationships, enabling accurate and predictive MPPT control. The predicted optimal operating point is used to control a Landsman DC–DC converter, while a PLL-based three-phase inverter ensures grid synchronization and power quality. The proposed controller was implemented in a HIL environment to verify real-time feasibility, and the performance results presented in this paper are based on both simulation and HIL validation. Simulation and Hardware-in-the-Loop (HIL) validation outcomes validate that the proposed method achieves a tracking efficiency of 99.2
A robust and efficient charging infrastructure is necessary due to the growing popularity of electric vehicles (EVs). A solar photovoltaic (PV)-based EV charging station's energy-efficient multi-level inverter technology is suggested by this manuscript. The Improved Spider Wasp Optimizer (ISWO), an optimization method enhanced by Beluga Whale Optimization (BWO), and the Evolutionary Gravity Neocognitron Neural Network (EGravity-NCNN) are integrated into the system. The goals are to lower switching losses, limit harmonics, and enhance the inverter system's efficiency and power quality. To minimize reliance on the grid, the ISWO algorithm is used to optimize solar energy usage and schedule EV charging sessions. A deep learning-based prediction model is created to estimate solar energy generation and EV charging demands. MATLAB simulations are used to assess critical performance indicators, including total harmonic distortion (THD), charging time, and operating efficiency. The simulation results proved that efficiency improvement in EV charging using this system was around 25
Balancing technical details and computational complexity is an important trade-off in long-term hydropower scheduling (LTHS) models. Rapidly increasing shares of variable renewable energy sources (VRES) challenges historical assumptions regarding the appropriate level of technical details and the representation of uncertainties in LTHS models. This work expands on previous research and formulates an efficient solution strategy to accommodate short-term variability and uncertainty into LTHS models, through gradually approximating the short-term dispatch problem per stage and under VRES uncertainty in the context of multi-stage Benders decomposition. The proposed solution strategy is embedded in an LTHS model based on stochastic dual dynamic programming and applied to a 2050 scenario of the Central and Northern European power system. We extract water values from the LTHS strategies and analyze their dependency on short-term variability.
India has a huge potential for harnessing solar energy. Expensive solar water heaters: High cost limits the adoption of solar energy systems, especially in rural areas. Affordable solar water heating systems for rural areas, lower energy consumption and carbon emissions; contributing to India’s renewable energy goals. India has an abundance of daylight. Sun-oriented energy is an ideal source of energy without any destructive outflow. Although there is a huge potential for harnessing solar-oriented energy in non-industrialized countries, the expensiveness of solar water heaters is limiting the growth of this market. In fact, it is common to use sunlight to heat water; however, it needs to be adjusted frequently to make better use of it. In this research, the researcher developed a polycarbonate sheet board with water supply channels and stage change material (PCM) for thermal power storage space. The experimental setup model were built and tested at UIT-RGPV, Madhya Pradesh, India. Taguchi procedure and ANOVA analysis were used to optimize the system performance. The component response diagram shows that the ideal element levels are selected as A1, B1, C1, D2, E1, F2, and G1. The saver tube is made of CPC material, the endothermic cover is made of polycarbonate, the number of collector tubes is 3, the distance of collector tubes is 1.27 cm2, the form of incorporation film is matte finished black brushed painting, the thickness of base intensity protection material is 12 mm, and the PCM is 46°. As displayed in the response diagram, the control variable F has the biggest impact on the productivity coefficient, trailed by B, G, A, D, C, and E, separately.
This paper introduces the design of a bidirectional grid-connected solar power electric vehicles (EVs) charging station (CS) with a focus on optimizing operational efficiency by minimizing grid power consumption. In this paper, a feed-forward Artificial Neural Network (ANN) is first employed to forecast the maximum power point (MPP) of a photovoltaic (PV) array using an extensive training dataset. To optimize the ANN training process, an adaptive Particle Swarm Optimization (PSO) algorithm is applied, achieving a tracking efficiency of 99.47
The rapid decline in fossil fuel reserves due to tremendous advances in technological and fiscal needs is inevitable, and hence emissions. Implementing demand side management (DSM) strategy in smart electric grids ensures the potent exploitation of sustainable energy resources and optimizes consumers’ consumption patterns while enhancing the grids’ stability. This paper proposes an optimal day-ahead load shifting DSM approach for a smart distribution grid comprising three load sectors, viz. residential, commercial, and industries. The objective function is mathematically formulated as an elementary quadratic minimization problem to achieve lower peak load and cost of operation. To optimize the minimization function, this paper proposes a hybrid optimization approach that combines class topper optimization (CTO) and modified Grey Wolf optimization (mGWO) algorithm to address the challenges of energy dispatch and load balancing in microgrids. The CTO algorithm provides robust global search capabilities to explore complex solution spaces. At the same time, mGWO ensures precise local convergence to optimize specific parameters such as energy cost, emission levels, and renewable resource utilization. The hybrid method leverages the strengths of both algorithms to overcome the limitations of standalone approaches. The proposed hybrid algorithm has significantly reduced peak load by 27.27
The increasing penetration of variable renewable energy sources has created significant operational and economic challenges for power systems. In Brazil, DESSEM – the official short-term generation planning model – relies on deterministic forecasts which, due to the uncertainty associated with wind power, often differ from real-time generation. These deviations lead to rescheduling actions that require the use of flexible hydroelectric and thermal units, directly affecting operating costs and day-ahead prices. This work proposes a two-stage stochastic model that takes into account redispatch costs resulting from forecast errors in wind energy. Based on a day-ahead generation computed in the first stage, the second stage determines the impact of this decision in terms of generation adjustments induced by different wind scenarios. The qualitative utility of our approach is demonstrated by its application to a simplified hydrothermal configuration which, although not dealing with 0–1 variables, incorporates the main characteristics of DESSEM. Out-of-sample simulations show that the two-stage model not only ensures the feasibility of different wind configurations, but also effectively reduces redispatch costs and improves the allocation of system flexibility. These results are particularly relevant given current discussions in Brazil regarding the implementation of mechanisms able to value flexibility and to provide robust price signals in the presence of a high penetration of variable intermittent sources of energy.
The accurate forecasting of solar power generation are crucial for optimizing energy management, maintaining grid stability, and supporting renewable energy integration. This study proposes a transformer-based model for solar power prediction, using its self-attention mechanism to capturing long-term dependencies and temporal patterns in time-series data. The model’s performance is compared with four advanced neural architectures: BiLSTM, Attention-LSTM, CNN, and GRU. The Transformer model outperformed all other models, achieving the lowest Mean Absolute Error (MAE: 0.0025) and the highest R^2 value (0.9998), indicating its superior accuracy and robustness. This paper provides a detailed methodology, covering data preprocessing, model architecture, and hyperparameter tuning, along with the mathematical formulations that define each model.
The increasing integration of Intermittent Renewable Energy Sources (IRES) and the need for decarbonization are driving significant transformations in the expansion planning of electrical power systems. In this context, the literature highlights the need to integrate long-term strategic decisions with short-term operational constraints in planning models, while preserving high resolution across time, space, and techno-economic detail. This paper addresses this challenge by proposing a high-resolution optimization model for the integrated expansion planning of generation, storage, and transmission in electrical power systems. The proposed model combines high temporal resolution, a multi-node spatial configuration, and a Clustered Unit Commitment (CUC) formulation to account for key operational constraints while limiting computational cost. The practical applicability of the model in real-world contexts is demonstrated through a multi-scenario analysis focused on the Italian electricity system, including a 2021 baseline scenario and 2030 future scenarios differing in terms of carbon tax and IRES support level. The validation of the model outputs in the baseline scenario against the 2021 historical electricity generation mix results in a Mean Absolute Percentage Error (MAPE) of 0.8
This article extensively compares two distinct artificial intelligence methods: Deep Learning (Long Short-Term Memory-LSTM) and Machine Learning (Piecewise ARX model-PWARX). The focus is on their effectiveness in understanding operational modes and how identify the discrete states within heating systems, particularly in solar and geothermal applications. PWARX proves adept at recognizing and comprehending various operational modes (discrete states), providing a detailed insight into system functionality and aiding in anomaly detection. In contrast, LSTM primarily serves as a validation tool, confirming established patterns within the data but lacking the deep physical understanding of operational modes. The article underscores PWARX’s strengths in system analysis and anomaly detection, highlighting its applicability in solar and geothermal heating systems.
This study develops a sequential two-pass lexicographic mixed-integer linear programme (MILP) for the decarbonisation of an off-grid phosphate complex in Saudi Arabia. The baseline comprises approximately 21 MWe of continuous electricity demand and 115 MW _th of dryer heat, supplied by diesel generation and heavy fuel oil (HFO), respectively. The model co-optimises photovoltaic generation (PV), concentrating solar power with thermal energy storage (CSP-TES), battery energy storage (BESS), and PEM electrolysis to replace diesel-based electricity and substitute HFO with green hydrogen. Pass 1 minimises annual unserved critical site electricity; Pass 2 minimises annual net cost over the reliability-feasible set, preserving a security-first design hierarchy without weighted objective trade-offs. Inter-annual solar uncertainty is represented through a Conditional Value-at-Risk (CVaR) extension using hourly NASA POWER irradiance data for 2013–2022. The site-only formulation applies α =0.95 , while the full site-plus-hydrogen formulation applies α =0.80 . Across the deterministic campaign, critical site electricity is fully protected in all cases. Under commercial-style financing, KAPSARC2030 remains in a low-investment partial-substitution regime, whereas Case2050 shifts to a high-investment design with approximately 1.24 BUSD of CAPEX, 19.74 kt yr−1 of hydrogen production, and 68.13 GWh yr−1 of residual unmet hydrogen service. Activating the hydrogen-service penalty raises production to 21.15 kt yr−1 and reduces unmet service to 1.07 GWh yr−1. Representative gross hydrogen cost is 3.85−4.39 USD kg−1 under a 2
This paper introduces a comprehensive dynamic reliability assessment framework for doubly-fed induction generator (DFIG)-based wind power systems, considering wind-speed-dependent component hazard rates. A bottom-up physical failure model is constructed by investigating the individual, operationally stressed failure trajectories of the key subsystems, including the blade assembly, generator, power electronics converters, transformer, and transmission cables. The results show that the equivalent failure rate is significantly modulated by the power loading profile of the turbine, with increased failure risk in the 10–15 m/s wind speed range due to maximum electro-thermal and mechanical stresses. Hourly availability and adequacy indices are quantified using a sequential Monte Carlo simulation over a one-year operational horizon. The proposed dynamic framework is rigorously evaluated against a traditional constant-failure-rate (CFR) baseline. The traditional method yields an optimistic LOLE of 3410.28 h/year, whereas the proposed wind-speed-dependent model indicates a non-conservative estimation error of 45.27
Reconfiguring conventional distribution systems into islanded networked microgrids enhances system resilience against severe power outages. However, maintaining stable operation during islanded conditions requires rapid and intelligent reconfiguration under uncertain operating environments. Key challenges include frequency regulation, distributed resource coordination, and real-time decision-making. To address these challenges, this paper proposes a dynamic coordination framework based on Double Deep Q-Networks integrated with Convolutional Neural Networks. The proposed approach explicitly models frequency variations and employs an exponential epsilon-greedy strategy to improve learning efficiency and convergence stability. A modified backward/forward sweep algorithm is used for islanded power flow analysis, while a graph-based Breadth-First Search method enables feasible topology reconfiguration. Extensive simulations demonstrate that the proposed Exponential Epsilon DDQN achieves up to 93
As a result of the legislation for gas markets introduced by the European Union in 2005, separate independent companies must conduct the transport and trading of natural gas. The current gas market in Germany, which has a market value of more than 54 billion USD, consists of Transmission System Operators (TSO), network users, and traders. Traders can nominate a certain amount of gas anytime and anywhere in the network. Such unrestricted access for the traders creates a free market while, on the other hand, it increases the uncertainty in the supply management and gas network operations. Some customers’ behaviors may cause abrupt structural changes in the gas flow time series. For this reason, it is challenging for the TSOs to predict the multiple hours-ahead gas nominations accurately. Our study aims to investigate the customers’ behavior in giving the nominations in advance for particular hours and to predict the final gas nominations up to 8 h ahead as precisely as possible. We propose an Automated Model Switching framework (AMS) for an accurate, robust, and efficient multi-step ahead prediction of entry point nominations in gas transmission networks. The results demonstrate that AMS achieves excellent performance, outperforming the best individual state-of-the-art models for the vast majority of the test cases while keeping the calculations as simple as possible.
A study on power system disturbance (PSD) classification of a distributed generator (DG) based grid-connected network has been carried out in this article. A total of 24 different PSDs commonly occurring in the DG-based grid-connected network have been considered. The three signal processing tools, Discrete wavelet transform (DWT), Detrended Fluctuation Analysis (DFA) and Recurrence Quantification Analysis (RQA) have been applied to construct a feature matrix. A category-wise visual representation on a 2D plane has been provided using the Uniform Manifold Approximation and Projection (UMAP) technique, and the coefficients of the feature matrix have been processed using this UMAP technique. The non-linear model, obtained from UMAP, helps to reduce the dimension of the classification algorithm and provides a categorical projection in a 2D plane. The category-wise visual 2D representation clearly proves the effectiveness of the model in identifying and classifying each class of PSDs. This huge variety of disturbance classification by visual 2D representation using UMAP is a novel approach in PSD study.
Short-term electrical load forecasting (STELF) is one of the crucial aspects of energy management systems, which helps in proper resource allocation and grid stability. In this paper, a novel approach has been proposed which leverages the advanced capabilities of the GRU model. This model is well known for its strength in capturing temporal dependencies of sequential data. Further, this approach is incorporated with the Class Topper Optimization Algorithm, which is a nature inspired meta-heuristic optimization process. The predictive capability of CTO-GRU model is enhanced by their integration through minimizing validation error via a time-ordered evaluation procedure to reach global optimal weight matrices. Further, the validity of the proposed framework is tested on real-time data sets through comprehensive experimentation. The resulting values demonstrate a significant gain in the accuracy, which justifies the better predictive capability achieved by the proposed approach. Through a proper analysis of the results, it can be assured that the proposed approach is suitable for enhancing the performance of the STELF. This research also provides opportunities for further improvement in the load forecasting methods by showing the advantages of combining state-of-the-art machine learning (ML) models with creative optimization algorithms to solve difficult real-world problems.
Cold Ironing (CI) is a proven strategy for reducing ship emissions at berth; however, its large, stochastic electricity demand creates significant technical and economic stress on port energy systems when supplied exclusively by the utility grid. This study proposes a smart sizing framework for a grid-connected Hybrid Renewable Energy System integrating photovoltaic and wind generation to sustainably supply CI operations. Using real operational data from a Mediterranean port, a high-resolution energy model is coupled with a Genetic Algorithm–based capacity optimization to determine the optimal renewable mix that minimizes both the levelized cost of energy and the carbon footprint of shore-side electrification. The resulting hybrid system is designed to maximize on-site renewable penetration while using the grid only as a balancing resource under load and resource uncertainty. Results show that the optimized configuration substantially reduces grid dependency and delivers major emission abatement compared with both grid-only CI and auxiliary-engine operation. The study demonstrates that smart capacity sizing, rather than real-time dispatch control, is the critical enabler of techno-economic and environmental viability for renewable-powered CI, and it provides a scalable, process-oriented decision-support framework for the design of sustainable port energy infrastructures. Development of a Genetic Algorithm–based smart sizing framework for hybrid photovoltaic–wind systems tailored to Cold Ironing demand profiles. Integration of real vessel traffic data and site-specific renewable resources to derive an optimal capacity mix that balances technical reliability with economic efficiency. Demonstration that the smart-sized hybrid system produces over 57 GWh/year and reduces grid dependence by more than 54
