This article introduces an improved regulation strategy for DC wind power systems linked to an imbalanced bipolar DC micro-grid. The study focuses on two distinct wind systems, each one connected to a separate terminal of the bipolar DC micro-grid and tested with varying wind speed profiles. The investigated wind systems are based on doubly fed induction generators with two voltage source converters: a rotor-side converter and a stator-side converter. The first one converter ensures the Maximum Power Point Tracking, while the second converter maintains balanced three-phase voltages on the stator side. Both converters employ predictive control algorithms. Simulations conducted using PSIM software prove the efficacy of the suggested control strategy in mitigating the impact of bipolar DC micro-grid imbalances on DC wind turbine efficiency. The system demonstrates performance even under challenging conditions, with DC bus variations reaching up to 50 %. To verify the system's precision and rapid dynamic response, the suggested control is compared with conventional resonant control through real-time simulations. A comparative analysis of the results demonstrates the superior suggested control efficiency for the studied system.
This study addresses the critical issue of overheating in built environments. With increasing extreme heat events and dense urbanization, overheating poses serious risks to human health, productivity, and well-being. Existing research mainly emphasizes external environmental drivers, leaving a gap in understanding internal building dynamics. This paper examines the thermal and energy behavior of an overheated educational laboratory room, pursuing two objectives: defining indoor comfort temperatures to enhance productivity and evaluating energy consumption to identify potential savings. A detailed numerical model of the laboratory envelope, incorporating interior electrical loads and air-conditioning systems, is developed using data from a monitoring setup that captures dynamic temperature variations. By applying the Narxnet predictive algorithm to historical temperature and humidity data, the model estimates energy consumption with high accuracy. Key challenges included creating a reliable model that reflects the complexity of the space, building a robust prediction method, and accurately representing airconditioning use. The technical approaches adopted to address these issues are presented, offering insights applicable to other overheated buildings. Results extend beyond the specific case, contributing to broader strategies for mitigating overheating. Validation confirms strong performance, with a 2% prediction error and less than 1 degrees C deviation in the envelope model, except for the glass fa & ccedil;ade, which reaches 1 degrees C. Estimated cooling demand also confirms the adequacy of the installed photovoltaic system, even under curtailment conditions, ensuring reliable and efficient operation.
In this study, numerical simulations were conducted to analyze the three-dimensional two-phase water/oxygen flow in the anode side flow field plate of a proton exchange membrane electrolyzer cell (PEMEC) using COMSOL Multiphysics at various times ( 0 s, 1 s, 2 s and 10 s). The mixture model was employed to capture the behavior of the two phases, aiming to investigate the flow characteristics within the flow field plate. Therefore, this software is used to solve numerically the complete three-dimensional model with the governing equations of continuity, momentum, and energy. The numerical findings including the velocity magnitude, the gas volume fraction distributions, and the pressure drop in the cell are presented and discussed. It is found that at 2 s and 10 s, the volume fraction of oxygen gas is highest at the plate centre of the flow field and increases from the channel entrance to the channel exit. Moreover, the flow distribution within the anode side flow field plate exhibits nonuniformity initially, particularly when gas production starts, but gradually stabilizes after reaching full oxygen production, indicating a relatively stable flow distribution between 2 s and 10 s. This work constitutes a contribution to the understanding of flow dynamics and the gas distribution within the anode side flow field plate of a PEM electrolyzer, crucial for optimizing its performance and efficiency.
This paper presents a nonlinear backstepping control design for a grid connected Photovoltaic (PV) system. This controller is used for a DC-DC boost converter to control the output voltage of the PV array according to a reference voltage produced by the Incremental Conductance (IC) algorithm. It is also applied to control the three phase inverter grid side to manage the DC bus voltage, active and reactive powers. The proposed controller is based on Lyapunov function to ensure the stability and to increase the effectiveness of the PV system. Simulation results are obtained in a Matlab/Simulink environment. The results indicate an excellent rapid response and fewer fluctuations even under the varying environmental conditions. It offers also an optimal PV power and a significant quality of current injected into the grid. Furthermore, it confirms the robustness in terms of injecting maximum active power into the grid, while keeping the DC bus voltage and the reactive power fixed at certain values. The validity of the Backstepping Controller (BSC) is proved with regard to extracting the Maximum Power Point (MPP) which is compared to the classic IC algorithm. Moreover, it is also demonstrated in the matter of controlling the Pulse Width Modulation (PWM) of the three phase inverter in comparison to the classic Proportional Integral (PI) controller.
The core contradiction in energy scheduling for new energy vehicles arises from a mismatch between the high-dimensional nonlinear characteristics of the energy system under complex dynamic driving conditions and the dual requirements of computational efficiency and accuracy in real-time optimization scheduling. To address this issue, this paper presents an energy scheduling algorithm based on continuous-time differential equation modeling within a model predictive control framework. A continuous-time differential equation model incorporating battery SOC (State of Charge) dynamics, vehicle dynamics, and regenerative braking is established, and an MPC (Model Predictive Control) framework is embedded to construct a physically constrained finite-time optimal control problem. Efficient numerical discretization and solution strategies are employed to achieve real-time optimization. Simulation and hardware-in-the-loop testing show that under four typical operating conditions, the algorithm's energy consumption per 100 km is 15.2 to 21.5 kWh/100 km, lower than regular strategies, classic MPC, and DQN (Deep Q-Network). The SOC change rate is reduced to between 0.35% and 1.25%, meeting the millisecond-level real-time requirements of the onboard system. This method combines physical interpretability and engineering deployability, and has application value in intelligent electric vehicle energy management, V2G (Vehicle-to-Grid) scheduling, and personalized energy-saving driving systems.
Given the increasing worries surrounding the impact of fuels, nations are now shifting towards sustainable energy production options through the utilization of renewable energy sources. In tandem with advancements in electric vehicle technology, the automotive production market is moving towards the electrification of transportation, contributing to cleaner cities. The integration of renewable sources, batteries, and electric vehicles into the grid leads to the utilization of bidirectional DC-DC converters to secure a two-way power flow. These converters come in various topologies and employ different control techniques based on their application domains. In this review paper, a thorough analysis is conducted concerning the advantages, limitations, and applications of various topologies of DC-DC bidirectional converters. The bidirectional DC-DC converters' main classifications are the isolated and non-isolated types. Within these categories, various topologies are compared. In addition, this review discusses novel topologies and emphasizes their contributions to the existing ones. Finally, an investigation is made into different control strategies utilized in the DC-DC bidirectional converters. This review aims to assist researchers by examining different configurations of DC-DC bidirectional converters. This will serve as a basis for comparing new designs selecting the most suitable converter for a particular application.
This paper presents a novel fault protection scheme for standalone low-voltage (LV) DC microgrids (MG), which relies on the first-order current and voltage derivatives to detect pole-to-pole (P2P) and pole-to-ground (P2G) faults in the system. Local measurements of current and voltage signals are applied for the proposed protection scheme. To effectively adapt to high noise levels of current and voltage sensors from the local measurements, a Chi-square-based statistic method is developed to determine tripping thresholds of the current and voltage derivatives in this proposed protection scheme. To be highly adaptable to the directional change of fault currents in the DC microgrid, moreover, the standard deviation- and mean-based calculation of lower and upper boundaries of current and voltage parameters has been also applied for the protection system. As a new contribution of the paper, this novel statistical and adaptive current- and voltage-based protection system can be more effective in protecting source and load branches of the small-scaled LVDC microgrid. Specifically, irrespective of large transients and measurement noises during the standalone operation of DC MGs, different P2P and P2G faults can be detected and cleared within a few milliseconds. The protection design procedure for DC microgrids is also detailed in this paper. For high applicability, solid-state relays and microcontrollers are equipped for the off-grid 48VDC microgrid testbed to implement and validate the suggested novel fault protection algorithm by doing multiple staged-fault tests at different positions in this microgrid testbed.
Voltage stability is a critical and important issue in power grid operation. STATCOM, as an adaptive FACTS controller, provides limited range linear control for non-linear processes. The major objective of the study presented in this paper is to prove and highlight the accuracy of the novel proposed control strategy of STATCOM optimized by combining classical evolutionary algorithms. The approach adopted for the STATCOM is a composition of existing controllers, PI, and Fractional PI. The novel control strategy aims to propose controllers in different loops tuned by evolutionary algorithms. These selected methods are particle swarm optimization (PSO), genetic algorithms (GA), and a hybrid combination of PSO and GA, which were adopted to tune the studied STATCOM control parameters in different strategies. Simulation results of this article are obtained using the Simulink environment. These results demonstrate the accuracy of the optimized two PI-PIFOPI STATCOM (PI controller in the outer loops of DC and AC voltage loops and cascaded PI and FOPI in the internal current loop) in stabilizing the grid voltage. This optimization was adopted by minimizing the objective selected in this study, the function ISE (Integral of the Squared Error) of grid voltage. The robustness and the accuracy of the proposed STATCOM are proved by random variation of the studied system parameters.
Charcoal is an extremely relevant (valuable) product for managing waste from wood offcuts or residues, because it can be used in a variety of applications, adding value to the material. This study aimed to investigate and compare the carbonization of three different types of wood, that is, teak (Tectona grandis) wood, matoa (Pometia pinnata) wood, and merbau (Intsia bijuga) wood, using a carbonization reactor. Each wood type had an initial weight of 2.15 kg and was processed for 4 hours. The study records the carbonization results, fuel gas consumption for each wood type, and the maximum temperature achieved during the process. The research findings indicate that Tectona grandis wood produces 0.75 kg of charcoal, utilizing 1.6 kg of fuel gas and reaching a maximum temperature of 374.07 degrees C. In contrast, Pometia pinnata wood yields 0.85 kg of charcoal, with 1.2 kg of fuel gas consumption, and a maximum temperature of 428.77 degrees C. Intsia bijuga wood generates 1.2 kg of charcoal, utilizing 1.3 kg of fuel gas and reaching a maximum temperature of 284.89 degrees C. Based on the data, it can be suggested that Teak wood charcoal has the highest charcoal content among the three types of wood, with a percentage of 79.02%. Meanwhile, Matoa wood charcoal has the lowest charcoal content, with a percentage of 44.56%, indicating that Teak wood is efficient for charcoal production.
This study explores the determination of an appropriate photovoltaic (PV) capacity for charging electric vehicles (EVs) through a case study on the electrification of clinic shuttle services in Japan. The purpose is to evaluate the feasibility of sustaining EV shuttle services using solar energy alone, without dependence on grid electricity, under diverse solar conditions. Recent efforts toward energy efficiency and decarbonization have accelerated the integration of renewable energy technologies and EVs. In the healthcare sector, shuttle services for elderly and mobility-impaired patients are widely operated, and their electrification is increasingly regarded as an important measure for reducing environmental impact. However, EV operation faces challenges related to charging availability and the variability of PV energy generation. This study analyzes the feasibility of photovoltaic-to-vehicle (PV2V) EV shuttle operation under varying meteorological conditions using real-world shuttle operation data. Meteorological conditions are classified into three categories using percentile-based thresholds of daily solar radiation, and month-long charging simulations are conducted using state-of-charge (SoC) trajectories estimated from actual driving records of clinic shuttle services. In addition, the impact of operational strategies is investigated by modifying shuttle route assignments while keeping the number of EVs fixed to reflect actual clinic operations. By redistributing driving distances among vehicles, the effect of operational load balancing on charging feasibility and required PV capacity is evaluated. The results indicate that appropriate route reassignment can significantly reduce the required PV capacity; for example, under low-solar-radiation conditions, the required PV capacity is reduced by approximately 31%.
Various Computational Fluid Dynamics (CFD) studies on high-rise buildings and horizontal axis wind turbines models have been conducted independently over the past decades. However, neither study has addressed the validation of results from both models within the same work. The primary objective of this study was to validate CFD simulations of a high-rise building and a horizontal axis wind turbine by employing the Realizable k-epsilon and SST k-omega turbulence models, aiming to determine the model that exhibits the highest accuracy when compared with experimental data available in the literature. Initially, models for the building and turbine were developed. Subsequently, grid independence studies were performed for both models. Finally, numerical results from both models were compared using validation metrics, including Hit Rate (HR), Normalized Mean Square Error (NMSE), and Mean Square Error (MSE). Overall, the Realizable k-epsilon model achieved superior results (NMSE = 0.022) compared to the SST k-omega model (NMSE = 0.039) in predicting the flow pattern on the building rooftop. Conversely, in simulations of the turbine, the SST k-omega(MSE = 0.370) outperformed the Realizable k-epsilon model (MSE = 0.445). These findings suggest that for CFD simulations of both models, particularly in urban wind energy applications, the SST k-omega model can be effectively employed.
In coastal landscape design, the arrangement of small-scale wind turbines often generates environmental integration challenges, including ecological disturbance, spatial fragmentation, and visual conflict. This paper proposes a collaborative optimization framework that integrates multi-source sensing data with generative design to balance renewable energy performance and coastal landscape quality. The reported results are derived from a hybrid approach combining field measurements, computational fluid dynamics simulations, and virtual reality-based participant evaluations. By combining dynamic ecological simulation with a closed-loop mechanism of public aesthetic feedback, the framework aims to achieve functional symbiosis and visual harmony between wind energy facilities and sensitive coastal environments. In implementation, a dynamic three-dimensional ecological sensitivity field is constructed by fusing satellite remote sensing, UAV surveys, ecological monitoring data, and bird radar observations, forming rigid ecological constraints for design generation. Morphological parameters, including turbine type, hub height, spacing, and tilt angle, are iteratively adjusted based on performance feedback from computational fluid dynamics wind field simulations. Public aesthetic evaluations collected through immersive virtual reality scenarios are quantified and transformed into computable weights, embedded into an iterative generative design loop to guide form and layout evolution. Through multi-objective optimization, the proposed method refines design solutions that reduce ecological disturbance, maintain stable energy output, and enhance visual compatibility. Ecological, energy, and aesthetic assessments confirm the effectiveness of the proposed method, with detailed metric ranges provided in the results section.
A solar photovoltaic (PV) power plant plays a vital role in meeting the growing demand for sustainable electricity. The reliability of such systems depends heavily on the quality of PV modules. This study presents a two-year qualification test of PV modules conducted at the BRIN PV Laboratory in Indonesia. A total of 168 modules representing more than 14 designs from local and imported manufacturers were evaluated for installation in a 71 kWp solar power plant. Results showed that 43% of modules failed to meet the required standards, with local products contributing to 83.3% of failures. Major defects included wet leakage current (59%), cell metallization burn (12%), glass breakage (8%), back sheet delamination (8%), and other issues such as soldering defects and junction box malfunctions (8%). Some modules exhibited more than 5% degradation in maximum power output by the end of the study. These findings highlight the importance of strict quality control, proper raw material selection, and careful handling throughout manufacturing and deployment. The study recommends strengthening quality assurance protocols and improving material selection to reduce failure rates. Enhancing module reliability is essential for ensuring durable PV systems and supporting the global transition to clean and sustainable energy.
Desalination holds promise as a sustainable solution to address the increasing need for water amidst the shortage of fresh water resources. But the high energy cost of traditional desalination techniques emphasizes the need to switch to more environmentally friendly technologies, particularly solar energy and other renewable sources. In this study, the authors proposed an inverted absorber sand bed solar still (IASBSS) layer of sand placed beneath basin water in an inverted absorber solar still (IASS). By using sand as a sensible storage material, the overall productivity of the conventional IASS increases. The developed mathematical model of the proposed IASBSS has been experimentally validated. For a day in Raipur, CG, India (21.251 degrees 81.629 degrees E). The daily performance of the proposed IASBSS is 14.9% higher than conventional IASS. The depth of water above the sand directly influences the amount of distillate produced by the still. When the depth of water increased from 1cm to 3 cm, the daily distillate output produced by the proposed method still decreased by 49%. The use of sand in IASS is found to be economical option to improve the production output from solar still.
The long-standing separation of energy systems and visual aesthetics in current urban environmental design makes it difficult to achieve a unity of function and aesthetics in the spatial composition of photovoltaic (PV) facilities. This paper proposes a Photovoltaic Aesthetic Integration Design Method (PAIDM) based on dynamic light energy mapping and multiobjective optimization. A high-precision radiative transfer simulation of solar flux is performed by establishing a spatial distribution model of light intensity. A multi-objective genetic algorithm (MOGA) is applied to achieve the global optimization of the spatial layout and tilt angle of the components with power generation efficiency, surface reflectivity uniformity, and visual coherence as joint objectives. This study utilizes a pure simulation approach to validate the proposed algorithmic design framework. In order to maintain equilibrium between energy conversion and color coordination in the PV system, a spectral sensing mapping function is built concurrently to correct the material reflectivity characteristics. The annual average irradiance of the meridional wall is reported to be 438.6 W/m(2 )from the experiments, which is equivalent to an energy density of 763.2 kWh/(m(2)& centerdot;year). The visual consistency score of the single-crystalline silicon material improved from 0.68 to 0.86 after spectral correction. With the help of this study, photovoltaic integration in urban design has been redefined as the use of an active rather than a passive component for generating sustainable urbanism, where aesthetic and energy-saving can still be maintained.
In the angular position control of servo systems, the issue of ensuring the Convergence (CVG) of the Tracking Error (TE) to a small and limited bound, as well as the chattering issue, are two important challenges. This paper addresses this challenge by presenting a new controller from the Sliding Mode Control (SMC) family. In this new controller, a new Reaching Law (RL) is used, in which a function based on the Barrier Function (BF) is used. The issue is that the tracking target, under any initial conditions, eventually converges to a small and limited bound, which is adjusted by the designer. In addition, this RL is robust to Model Uncertainties (MU), providing a robust performance for controlling the position of this machine. The stability of this controller is proven by Lyapunov theory, during which the CVG of the TE to a limited bound is guaranteed. In addition, this controller has the ability to decrease the chattering phenomenon to a good extent. A series of practical tests was conducted in the laboratory to examine the performance of the Proposed Method (PM), and the results confirmed its effectiveness in the above-mentioned field.
Mounted Wind Turbines (BMWTs) are installed on building rooftops to exploit the wind velocity amplification found there. Optimal positioning of BMWTs (micrositing problem) can increase the energy gathered from the wind and reduce the total energy cost. Although micrositing methodologies have been extensively studied for wind farms, a gap in knowledge exists regarding the micrositing of BMWTs. The main objective of this work was to propose a methodology for optimal micrositing of BMWTs using Computational Fluid Dynamics (CFD) and Genetic Algorithms (GA). Thus, a site assessment was initially performed. Wind data treatment was carried out to determine those wind velocities and directions to be used in the next stages. These wind velocities and directions were simulated within an urban environment via CFD. The selection of a BMWT was then carried out. Furthermore, the zones with low wind speeds and high turbulence levels restricted the search space used in the GA-based micrositing optimization. Finally, a sensitivity analysis employing the building reinforcement factor (F) was performed. The results showed that the energy produced yearly by the BMWT is the key parameter in reducing the Cost of Energy (CoE), which achieved a value of 3.05146 $/kWh.
This study presents an advanced peer-to-peer energy trading framework that integrates renewable energy source uncertainty modeling, demand response, and multi-objective optimization using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The proposed approach aims to minimize carbon dioxide emissions while maximizing social welfare within a decentralized microgrid network. Uncertainties in photovoltaic irradiance, wind speed, and load demand are modeled using lognormal, Weibull, and normal probability density functions, respectively. A recurrent neural network is employed to forecast baseline day ahead renewable generation and demand profiles, providing accurate temporal data for optimization. The fitted distributions are then used to represent uncertainty around these baseline forecasts for scenario construction. The simulated system comprises 60 participants, including 20 prosumers and 40 consumers, equipped with photovoltaic units, wind turbines, battery energy storage systems, and electric vehicles. Two operational scenarios are examined: a baseline case without DR and an enhanced case including DR. In the baseline scenario, the optimal solution achieves a social welfare index of $6721.449 and total emissions of 3255.172 kg CO2. When DR is implemented, emissions are reduced by approximately 17.3%, while welfare decreases by about 14.6 %, revealing a clear environmental-economic trade-off. Pareto front analysis confirms that demand response participation effectively reduces demand peaks, improves renewable energy utilization, and lowers reliance on diesel generators. Overall, the NSGA-II-based P2P trading and DR coordination framework enhances sustainability, flexibility, and robustness under renewable uncertainty. The proposed model provides a scalable decision-support tool for emission-conscious and welfare-optimized smart grid communities, supporting efficient integration of distributed energy resources in low-carbon electricity markets worldwide.
