This research presents a techno-economic assessment of hydrogen production and storage systems (HPSSs) with solar photovoltaic (SPV) systems and a wind farm (WF) located at the Mamatkheda village in India. The proposed system seeks to satisfy the dynamic load demands of different settlements in the area under study. An optimal size of the HPSS is determined using the HOMER Pro software tool based on the geographical data, meteorological data, and monthly energy requirements of the village. The study considers three different configurations namely, base system, SPV-WF-biomass plant (BP), and SPV-WF-HPSS, for detailed study in HOMER Pro with 25 years of lifetime, 8

This paper presents a detailed modeling and simulation framework for a solar-fed electric ferry system based on a unified DC microgrid architecture integrating photovoltaic (PV) generation, battery energy storage, and electric propulsion. The proposed model captures the complex interactions between renewable energy variability, battery dynamics, and propulsion load under realistic operating conditions. A high-fidelity battery model is developed to characterize nonlinear discharge behavior, including state of charge (SoC) evolution, terminal voltage variation, and load-dependent current response. The system incorporates a bidirectional DC–DC converter to enable controlled charging and discharging, ensuring effective energy management and DC bus voltage regulation. Additionally, a closed-loop controlled DC motor drive is implemented to achieve stable and responsive propulsion performance under dynamic loading conditions. The integrated simulation framework evaluates system behavior under both transient and steady-state scenarios, demonstrating the ability to maintain power balance, ensure reliable operation, and efficiently utilize available solar energy. Results indicate that while the system effectively adapts to load variations, battery performance is strongly influenced by depth of discharge and internal characteristics, which directly impact voltage stability and overall efficiency. The proposed approach provides a comprehensive foundation for analyzing and optimizing solar-assisted electric ferry systems, contributing to the development of reliable, energy-efficient, and sustainable maritime transportation solutions.

Increase in trading activities and marine transportation has led to excess emissions at seaports during berthing periods, resulting in a growing adoption of electricity-based cold-ironing (CI) facilities. To reduce direct dependency on the utility grid for CI facilities, energy storage systems and renewable energy sources can be incorporated at seaports. A more viable option involves forming a seaport microgrid by integrating multiple shipboard microgrids (SMGs) through port-based charging infrastructure. However, power management for electric ships becomes complicated when a large number of ships start to integrate at seaports through charging stations. This paper proposes a grid-connected ship-based seaport microgrid with a fuzzy-based power management strategy for efficient power sharing in CI facilities as a solution. This approach is particularly beneficial for autonomous ships and islands where traditional port electrification may not be technically feasible. The proposed power management scheme is validated on the MATLAB® Simulink platform through extensive numerical simulations.

Accurate forecasting of Global Horizontal Irradiance (GHI) is critical for reliable photovoltaic (PV) integration and efficient grid operation, yet conventional statistical and machine learning models struggle to adapt under nonstationary and highly variable weather conditions. This paper presents a novel framework for Solar irradiance forecasting using Deterministic Policy Gradient-based reinforcement learning with asymmetric neural network architectures for actor and critic networks. The actor network employs a hierarchical convolutional module with multi-scale kernels to capture both short-term local variations and long-term temporal trends in solar irradiance data, while the critic network incorporates parallel fully connected layers to stabilize value estimation and reward optimization. The framework is evaluated on five years of GHI data (2010-2014) from three climatically diverse Indian locations—Kochi (coastal), Bengaluru (moderate elevation), and Manali (high altitude). In Kochi, our RL Agent reduced RMSE and MAE by 3.55% and 4.77%, respectively, compared to the second-best baseline; in Bengaluru, RMSE decreased by 2.64% and MAE by 5.78%; while in Manali, reductions of 2.39% (RMSE) and 4.35% (MAE) were achieved. The method also improved R2 scores across all sites, with gains of up to 3%. Furthermore, with previous GHI as an additional input, the DDPG RL-based framework again improves upon the baseline methods. These results highlight the robustness of our RL Agent with its novel asymmetric design and multi-scale feature extraction in diverse climatic conditions, establishing it as a promising solution for adaptive solar forecasting and enabling more reliable integration of renewable energy into power systems.

Proton Exchange Membrane Fuel Cells (PEMFCs) are key to facilitating the global shift towards renewable energy sources. Nonetheless, costs associated with purchasing expensive equipment/systems and the dangers involved with utilizing pressurized hydrogen impede progress in PEMFC research and education. To enable continued development of PEMFCs without the need for physical prototypes, we have created a complete Digital Twin of the PEMFC using the Software-in-the Loop (SIL) approach to correlate a high-fidelity Simscape model developed within MATLAB/Simulink, a custom-built backend using Python and a Stream lit visualization dashboard. The digital twin developed in this project allows for the simulation of thermodynamic and electrochemical behavior in real-time, incorporating a novel bi-directional control mechanism to facilitate active controllability of the Digital Twin via Python script-based updates on MATLAB workspace variables instead of merely passive monitoring. In addition, the system also integrates deterministic fault detection algorithms for detecting critical failure modes including gas leakage, voltage collapse, and thermal runaway. The results from this research validate the ability of the Digital Twin to accurately replicate the characteristics of physical polarization characteristics of actual PEMFC stacks and indicate that the Digital Twin represents a very cost-effective and safe alternative to hardware in the design and fault analysis phases.

This paper introduces an optimized energy management framework aimed at improving the efficiency of a tri-hybrid motorbike, specifically calibrated for the Chennai Motorbike Driving Cycle (CMDC), which reflects the real-world traffic environment of Chennai, India. The proposed hybrid configuration combines a fuel cell for steady energy generation, a battery for medium-term storage, and an ultracapacitor for handling rapid power transients ensuring effective response to the frequent start-stop and acceleration patterns of urban travel. To optimize the power-sharing strategy among these three sources, the study employs Adaptive Particle Swarm Optimization (APSO), a metaheuristic approach with dynamically tuned inertia weight. This method surpasses traditional rule-based control by continuously adapting power flow to reduce total energy consumption and extend the operational lifespan of individual components. Using the CMDC profile of 9.09 km distance and 22.7 km/h average speed as a validation benchmark, simulation outcomes confirm that the APSO-based strategy yields higher energy efficiency and enhanced durability compared to fixed allocation techniques. Moreover, sensitivity assessments highlight the framework’s resilience under variable conditions such as speed fluctuations, power constraints, and battery state-of-charge ranges. The research contributes toward sustainable urban mobility by presenting a scalable control solution for two-wheelers in traffic-dense cities like Chennai, supporting India’s transition toward cleaner, energy-efficient transportation systems.

A high-efficiency 20 kW electric vehicle charger was designed using a Dual Active Bridge converter with galvanic isolation to ensure safe and reliable power transfer. To maintain optimal performance across varying load conditions, a multi-modulation control strategy was implemented, incorporating Single Phase Shift, Dual Phase Shift, Triple Phase Shift, and Extended Phase Shift techniques. An adaptive logic loop was developed to dynamically select the appropriate modulation based on real-time power levels and error feedback. Zero Voltage Switching was continuously monitored and achieved across all switches, minimizing switching losses and enhancing system efficiency. The system was modelled and simulated in MATLAB/Simulink. Simulation results confirmed smooth transitions between modulation schemes, improved energy transfer, and stable operation across the full power range. The proposed methodology offers a scalable and effective solution for high-power EV charging applications

The rapid growth of electric vehicles has intensified the demand for efficient, reliable, and sustainable charging infrastructure. Conventional electric vehicle charging stations, predominantly dependent on grid power, face challenges such as peak load stress, high operational costs, and reliance on fossil-fuel-based electricity. To address these issues, this work proposes an optimized multi-stage charging strategy for solar-integrated, grid-connected electric vehicle charging stations equipped with Battery Energy Storage Systems. The system architecture enables intelligent power flow management among the photovoltaic array, grid, battery energy storage systems and multiple electric vehicles charging ports under varying load and generation conditions. A multi-step constant current charging method is implemented, which adapts charging current in electric vehicles based on battery state of charge to enhance charging efficiency, ensure safety, and extend battery life. An incremental conductance-based maximum power point tracking algorithm is employed for effective solar energy utilization. Simulation studies in MATLAB/Simulink validate the effectiveness of the proposed strategy, demonstrating improved photovoltaic utilization, reduced grid dependency, and stable operation. The results highlight the potential of solar-BESS integration and multi-step constant current charging to provide an energy-efficient, cost-effective, and environmentally sustainable solution for electric vehicle charging stations.

The increasing adoption of multi-source electric vehicle architectures has motivated the development of advanced energy management strategies capable of handling highly dynamic urban driving conditions. Tri-hybrid electric vehicles integrating a fuel cell, battery and ultracapacitor combine complementary energy and power characteristics but their effective operation depends on coordinated power sharing among the sources. This paper presents a fuzzy logic-based energy management strategy for a tri-hybrid electric vehicle developed and evaluated using MATLAB/Simulink. The proposed controller allocated traction power in real time based on instantaneous power demand and battery state of charge, allowing smooth fuel cell operation, controlled battery utilization and efficient use of the ultracapacitor for transient loads. A simplified power-based vehicle model is implemented to emphasize system level energy management behavior rather than detailed component electrochemical dynamics. The performance of the proposed strategy is assessed under the FTP-75 urban driving cycle. Simulation results demonstrate accurate power balance between demanded and supplied power, stable battery state-of-charge regulation within prescribed limits, clear role separation among the energy sources and effective power utilization under dynamic operating conditions. Energy contribution and utilization analyses confirm that the fuel cell supplies most energy demand, while the ultracapacitor supports high power transients, validating energy management for tri-hybrid vehicles.

The electrification of maritime transport requires coordinated management of multiple energy carriers to ensure operational reliability and economic viability. This paper presents a real time energy coordination strategy for an integrated electric ferry and port level multi energy system comprising renewable energy sources, hydrogen conversion and storage units, battery swapping infrastructure and grid interaction. The proposed approach employs a reinforcement learning based control mechanism to manage coupled electricity and hydrogen flows under variable renewable generation electricity price uncertainty and discrete transport operations with relaying on optimization solver. System dynamics are modeled using realistic operational parameters and both grid connected and islanded operating modes are considered. The methodology is evaluated under multi year scenarios representing 2025,2035 and 2045 to capture the impact of technological advancement and demand growth. Simulation results demonstrate stable electric ferry operation, balanced battery swapping station performance and progressive improvement in hydrogen storage utilization, while maintaining limited dependence in grid power. Learning convergence characteristics indicate consistent adaptation towards cost efficient operating policies. The results confirm the suitability of the proposed strategy for renewable dominant port energy system and highlight its potential to support future port electrification and maritime decarbonization initiatives.

The effective integration and operation of photovoltaic (PV) generation in power systems are critically dependent on the availability of reliable and accurate day-ahead global horizontal irradiance (GHI) forecasts. Traditional methods, however, face serious issues due to strong nonstationary and nonlinear GHI characteristics that are influenced by varying atmospheric conditions. To address these issues, this paper proposes a hybrid forecasting methodology that combines variational mode decomposition (VMD) and the convolutional neural network (CNN) for day-ahead GHI prediction. In the proposed method, the VMD approach is specifically applied to the GHI to reduce its non-stationary characteristics, while the temperature, RH, and clear sky GHI are used in their raw form to maintain their physical interpretability. The CNN model then uses the meteorological variables and the decomposed GHI components as inputs to identify significant temporal patterns. The VMD-CNN approach is validated using real-world scenarios with three different Indian Locations-Jodhpur, Kochi, and Manali-and tested using benchmark models such as RF, ANN, and standalone CNN. The results demonstrate consistent improvements across all locations, with the proposed technique achieving an average reduction of 33.52% in RMSE and 33.7% in MAE compared to the CNN model, along with high $\mathbf{R}^{\mathbf{2}}$ across all locations. The outcome of these results highlights the effectiveness and robustness of the presented framework in effectively capturing complex irradiance characteristics, which makes it suitable for real-world day-ahead solar forecasting applications.

Agricultural production requires careful management of inputs such as fungicides, insecticides, and herbicides to ensure a successful crop that is high-yielding, profitable, and of superior seed quality. Current state-of-the-art field crop management relies on coarse-scale crop management strategies, where entire fields are sprayed with pest and disease-controlling chemicals, leading to increased cost and sub-optimal soil and crop management. To overcome these challenges and optimize crop production, we utilize machine learning tools within a virtual field environment to generate localized management plans for farmers to manage biotic threats while maximizing profits. Specifically, we present AgGym, a modular, crop and stress agnostic simulation framework to model the spread of biotic stresses in a field and estimate yield losses with and without chemical treatments. Our validation with real data shows that AgGym can be customized with limited data to simulate yield outcomes under various biotic stress conditions. We further demonstrate that deep reinforcement learning (RL) policies can be trained using AgGym for designing ultra-precise biotic stress mitigation strategies with potential to increase yield recovery with less chemicals and lower cost. Our proposed framework enables personalized decision support that can transform biotic stress management from being schedule based and reactive to opportunistic and prescriptive. We also release the AgGym software implementation as a community resource and invite experts to contribute to this open-sourced publicly available modular environment framework. The source code can be accessed at: https://github.com/SCSLabISU/AgGym.

Many residences are adopting electric vehicles and installing photovoltaic systems on the rooftop of their house. Due to dynamic tariffs, it becomes complicated to manage the electricity. This paper develops a cost-effective and computationally efficient way to predict how to control the charge and discharge cycles of electric vehicles in smart homes with rooftop photovoltaic systems and changing tariff structures. The framework does not rely on complicated deeplearning architectures. Instead, it uses Holt-Winters exponential smoothing for short-term load and photovoltaic generation forecasting and trend-seasonal decomposition for electricity price prediction. This makes it possible to make accurate hourahead forecasts that can be utilized in real time. These predictions are used in a scheduling model based on linear programming that aims to lower net operating costs while still following battery state-of-charge limits, residential charger limits, bidirectional operation, and costs related to degradation. The suggested system uses real dataset to make it possible to charge at the same time when prices are low and solar output is high, and to discharge at the same time when prices are high. The results show a net daily profit of around Rs 177 and a big drop in reliance on the grid. This shows that using model predictive control-driven electric vehicle scheduling for home energy management is a practical option. The methodology offers a simple, scalable, and affordable alternative to optimization and deep learning-based approaches that require a lot of computation.

Open-circuit switch faults (OCSFs) in power semiconductor switches are caused by wire bonding failures, gate driver malfunction, surge voltage/current, electromagnetic interference, and cosmic radiation. Under OCSFs, the signal characteristics are not excessively high, but prolonged OCSFs risk cascading system failures. This letter presents a comprehensive analysis of various deep neural network (DNN)-based architectures, such as long short-term memory (LSTM) and convolutional neural network (CNN), to diagnose multiclass OCSFs in three-phase active front-end rectifiers (TP-AFRs). A novel multisensor time-series sequence (MTSS) dataset is acquired at 500 Hz, comprising 624 observations from 19 sensor signals for single, double, and triple-switch OCSFs. The intertwining issue in the MTSS dataset is visualized using t-SNE, and the initial experiments with support vector machine (SVM) rendered the highest test accuracy of 93% against k-nearest neighbor, artificial neural network, and decision tree classifiers. Further, our investigations revealed that an architecture with two-layer CNN, one-layer LSTM, and one fully connected layer achieves a competitive testing accuracy of 95.03%, showing an improvement of 2.03% from the SVM classifier, and 7.03% from the one-layer LSTM network. These findings demonstrate the potential of this approach for enhancing reliability of TP-AFRs with the direct application of downsampled raw electrical signals.

This research enhances residential electric vehicle (EV) charging by integrating maximum demand (MD)-based regulation, which improves grid management while minimizing infrastructure and maintenance costs. Multiple EVs connected to distribution transformers (DTs) during peak hours lead to voltage drops, increased losses, and unfair charging distribution across households. This article introduces a decentralized EV charging system that integrates MD-based regulation, considering local parameters such as: 1) power consumption of household loads; 2) grid supply voltage; and 3) state of charge (SOC) of the battery. The proposed approach maintains EV charging within the rated capacity of the DT and the maximum allowable load, effectively minimizing load variance while ensuring a continuous power supply to household loads. To validate the proposed system: 1) a hardware-in-loop (HIL) simulation is conducted with Opal-RT; 2) the responses under various loadings are tested in a Simulink model; 3) a comparison study conducted using real household consumption data; and 4) network integration test and comparison study conducted using modified IEEE low-voltage distribution network (LVDN) in Simulink. The results demonstrate that this approach minimizes load variance, reduces peak-to-average ratios, and achieves fair, reliable charging across the network, regardless of EV connection points, making it a scalable and efficient solution for residential EV integration.

The uncoordinated charging of electric vehicles can cause incremental overloads, power losses and voltage fluctuations, posing significant stress and risks to distribution networks. To mitigate these challenges, effective electric vehicle charging strategies are essential. This paper proposes a time of use pricing based optimization approach to minimize electric vehicle charging cost by optimizing the charging-discharging strategy and rate of charge and discharge while adhering to constraints such as state of charge limits, charge-discharge rate limit, desired departure state of charge and distribution feeder capacity. To address uncertainties in arrival time, departure time and initial state of charge of electric vehicles, normal distribution probability functions are applied. The proposed method is tested on residential distribution network with electric vehicle penetration including and excluding solar photo-voltaic. Using mixed integer non-linear programing for optimization, the results demonstrate that plug-in electric vehicles are incentivized 25.63% more when powered by a solar PV system compared to a system without solar PV. The uncoordinated charging leads to a high expenditure but the optimized approach transforms the operation into profit making scenario.

Electric vehicles (EVs) are revolutionary mobility solutions to limit environmental pollution. Increasing EVs negatively impacts power distribution system performance and power supply quality. A significant impact is on the distribution system network, due to which the power supply's quality deteriorates, resulting in increased distribution and transmission losses. As a result, the overall efficiency of the power distribution network is reduced. This study analyzes and discusses the impact of EVs on the 34-bus radial distribution system. In this work, the smooth load curve is achieved by reducing load variance. Here, convex optimization method is used to solve the load variance problem. Further, the improvement in the performance of the distribution system network due to the integration of EVs is investigated. The intended goal is accomplished by connecting the Open DSS software with MATLAB software. For analyzing the various data, Python software is used. Ultimately, this paper improves the load profile and voltage profile by minimizing the load variance.

The article proposes a high-performance control approach for a three-phase single-stage photovoltaic (PV) inverter to feed only active power into the grid and compensate load reactive power demand in grid-connected mode. The key objective of this article is to propose a second order sliding mode control (SOSMC) with modified super twisting algorithm (STA) for the PV inverter to regulate DC link voltage under varying weather and load conditions. The injected grid current is also adjusted to ensure unity power factor operation, thus reducing line losses. The modified STA is designed with boundary conditions to reduce chattering. A conventional algorithm is employed to extract the maximum generated solar power from PV array. The system performance with modified control method is assessed by assessing the obtained results with conventional STA-based SOSMC and PI-based voltage-oriented control. Extensive numerical simulations are performed, and the test results with proposed control method are presented to validate operations of the grid-connected PV inverter. The results have demonstrated the effectiveness of the modified STA in maintaining the DC link voltage under varying weather, loads, and fault conditions, based on standard error indices.