Professor Farhad Shahnia

Surface anomalies such as dust accumulation and bird droppings on photovoltaic (PV) panels can significantly reduce their energy production and lead to inefficient maintenance decisions. This paper proposes a vision-based deep learning framework for the automatic detection of PV panel surface conditions and validates the detected anomalies using real inverter-level energy production data. Unlike conventional studies focusing solely on detection performance, the proposed approach introduces a unified and physically interpretable framework that directly links image-based anomaly detection with inverter-level energy performance and decision-oriented PV maintenance. An EfficientNetB3-based model is trained using a two-stage transfer learning strategy on a publicly available Kaggle dataset and evaluated using standard classification metrics. The trained model is then deployed and validated at a 1 MW solar power plant located at Karaman, Türkiye. Classification results obtained from field images are systematically linked with inverter-associated hourly energy production measurements. Following panel cleaning and natural rainfall, an approximately 12.5% increase in inverter-level hourly energy production is observed for the analyzed PV group (120 panels, ~270 Wp), corresponding to an increase from 23.2 to 26.1 kWh. In addition, the study introduces an energy–water–sustainability-aware cleaning decision framework tailored for arid and semi-arid regions where water scarcity and deep groundwater extraction present critical constraints. The framework defines a quantitative decision rule in which panel cleaning is performed only when the expected recoverable energy exceeds the energy cost of water extraction and cleaning. Overall, the proposed approach enables accurate surface anomaly detection while supporting sustainability-aware, resource-efficient and data-driven maintenance decisions for PV power plant operation.

Onshore, fixed-bottom offshore, and floating offshore wind turbines (FOWTs) represent distinct technological pathways for large-scale wind energy deployment, yet comparative assessments often emphasize descriptive characteristics rather than quantified performance trade-offs. This study presents a structured comparison of these three wind energy systems across technology maturity, levelized cost of energy (LCOE), environmental impacts, and socio-regulatory constraints. A harmonized literature synthesis of recent techno-economic data (2018–2024) reveals that onshore wind remains the lowest-cost option (typical LCOE: 30–55 USD/MWh), while fixed-bottom offshore wind exhibits higher but declining costs (60–100 USD/MWh) driven primarily by installation and foundation expenses. Floating offshore wind systems currently show the highest LCOE (90–160 USD/MWh), dominated by platform and mooring costs, yet demonstrate the strongest long-term cost-reduction potential due to access to superior wind resources and deep-water sites. Environmental and social analyses indicate increasing public acceptance challenges for onshore projects, while offshore and floating systems face regulatory complexity and higher capital risk rather than social opposition. The study’s contribution lies in a balanced, criterion-based framework that exposes cost drivers, maturity gaps, and deployment constraints across wind technologies, highlighting floating offshore wind as a transitional technology whose competitiveness depends on platform standardization and supply-chain learning effects. The analysis demonstrates that cost parity between floating and fixed-bottom offshore wind is not turbine limited but platform- and financing-driven, with regulatory clarity having a comparable impact to capital expenditure (CAPEX) reductions.

In light of the growing demand for sustainable energy and the diminishing viability of conventional fossil fuels, wind energy has become a compelling alternative particularly in wind-rich regions such as Morocco. Among the various variable-speed wind turbine technologies, the Doubly Fed Induction Generator (DFIG) stands out for its efficiency and cost-effectiveness. However, traditional control methods for DFIGs, including Field-Oriented Control and Direct Power Control (DPC), are hindered by limitations such as sensitivity to parameter variations, torque and flux oscillations, and fluctuating switching frequencies that contribute to harmonic distortion. To address these challenges, this study introduces an enhanced power control strategy for DFIG-based Wind Energy Conversion Systems (WECS), incorporating the Bald Eagle Search (BES) algorithm into the FOC framework. The proposed method dynamically adjusts the reference values of active and reactive power in real time, guided by system performance metrics such as power error and flux stability. Inspired by the bald eagle’s hunting strategy, the BES algorithm enables rapid convergence, high adaptability, and reduced computational complexity. Simulation results validate the effectiveness of the proposed BES-based controller, showing superior performance compared to conventional FOC and Fuzzy Logic Controller methods particularly in terms of harmonic distortion and dynamic response. The Total Harmonic Distortion (THD) values achieved are 2.69% for FOC, 2.52% for FLC, and 2.51% for the BES-based method, highlighting its potential to enhance power quality and operational stability under varying wind and grid conditions.

This paper presents a novel approach to optimizing vehicle-to-grid (V2G) enhanced energy management in microgrid systems through machine learning-based forecasting. The proposed system utilizes the Random Forest algorithms to predict energy consumption and renewable energy generation patterns, enabling intelligent decision-making for V2G operations. The proposed methodology incorporates temporal features including hourly, daily, and monthly patterns to create accurate 24-hour forecasts for both load demand and renewable energy generation. The developed V2G optimization strategy then uses these forecasts to make informed decisions about the charge and discharge timing of electric vehicles, maintaining a balance between immediate grid requirements and anticipated future needs. The performance of the proposal is evaluated using real-world microgrid data and demonstrates significant improvements against traditional V2G management approaches. The studies demonstrate that the proposed model uses battery cycles more efficiently, prevents unnecessary energy transfers, and reduces battery degradation by minimizing excessive charging. The numerical studies show that the proposed technique maintains the energy deficit at a lower rate by 6.4% fewer charge-discharge operations. Furthermore, the proposed approach relies on easily accessible data rather than difficult-to-obtain weather variables, enhancing its practicality and ease of implementation. This makes the system more applicable in real-world scenarios without requiring complex meteorological data collection and enables the proposed system to adapt to varying renewable energy generation patterns and consumption behaviors, particularly suitable for microgrids with high renewable energy penetration. This research contributes to the growing intelligent energy management systems field and provides a practical framework for implementing machine learning in V2G applications.

This paper introduces a new four-switch, high-voltage, high-step-up converter employing two transformers. The topology enables Zero-Voltage Switching (ZVS) across all primary switches for operating conditions ranging from no load to full load. A voltage-quadrupler and a voltage-doubler rectifier are used on the secondary sides of the transformers, enabling reduced turn-off current for the voltage-quadrupler diodes and Zero-Current Switching (ZCS) turn-off for the voltage-doubler diodes, thereby ensuring high efficiency across diverse load levels. Notably, the voltage stress experienced by the voltage-multiplier diodes is significantly lower than the output voltage, thereby rendering the converter exceptionally suitable for high-voltage applications such as electron beam welding (EBW). The voltage gain surpasses that of the conventional phase-shift full-bridge (PSFB) converter, permitting a lower transformer turns ratio and thus reducing winding resistivity. The removal of the substantial output inductor leads to a lighter and more compact design, eliminating insulation concerns associated with inductor windings. This paper details the operation of the proposed converter, supported by experimental results from a 500-W prototype with a 150-V input and 2-kV output, which confirm its high performance and operational advantages.

Uncertainty in renewable energy resources and variations in the demand response (DR) of participation pose significant challenges for accurately predicting participation levels, particularly across seasons. This study offers a holistic approach to seasonal DR scheduling within the AC optimal power flow (AC-OPF) framework, focusing on advanced thermal energy storage (TES) and optimal energy storage system (ESS) allocation. In this study, we categorized the electrical loads and energy generation from solar and wind sources on a seasonal basis. By analyzing historical load data using the K-means clustering method, we identified key scenarios for implementing effective seasonal demand response (DR) strategies. Additionally, we forecast the optimal participation rates for subscribers, enabling the design of targeted DR incentives that meet the seasonal system needs. Our research specifically targets the optimization of heating and cooling demands within HVAC systems, emphasizing the impact of ambient temperature on the efficiency of TES. This relaxed mixed-integer nonlinear programming (RMINLP) problem was solved using GAMS software with the CONOPT3 solver, specifically applied to the IEEE 24 and 118 bus networks. This multi-objective optimization framework enhances energy management and facilitates the integration of renewable resources, thereby contributing to grid stability and sustainability. Our findings highlight the crucial role of seasonal DR strategies for enhancing the overall efficiency of energy systems. For each season, an optimal range and point for DR performance was obtained to be offered to subscribers by the system operator.

Optimal scheduling of integrated PV/wind energy systems (IESs) is a complex task that requires innovative approaches to address uncertainty and improve efficiency. This paper proposes a novel multi-objective optimization framework for IES operation, incorporating demand response (DR), a comprehensive set of components, and innovative techniques to reduce computational complexity. The proposed framework minimizes total losses, cost, and emissions while meeting energy demands, offering significant advantages in terms of sustainability and cost reduction. The optimization model is implemented using steady-state energy analysis and non-dominated sorting genetic algorithm-III (NSGA-III) heuristic optimization, while uncertainty analysis and scenario reduction techniques enhance computational efficiency. To further reduce the computational burden, the proposed framework incorporates a novel clustering strategy that effectively reduces the number of scenarios from 1000 to 30. This innovation significantly improves the computational efficiency of the proposed framework, making it more practical for real-world applications. The effectiveness of the proposed approach is validated against multi-objective seagull optimization algorithm (MOSOA)- and general algebraic modeling system (GAMS)-based methods, demonstrating its superior performance in various scenarios. The improved management system, enabled by the proposed algorithms, facilitates informed operational decisions, enhancing the system’s installed capacity and overall flexibility. This optimization framework paves the way for more efficient and sustainable operation of integrated PV/wind energy systems. Reducing gas and heat network losses, considering both electric and thermal load response, simultaneously utilizing electricity, gas, and heat storage devices, and introducing a new clustering strategy to reduce scenarios are the specific innovations that are mentioned in this paper.

The increasing adoption of electric vehicles (EVs) has led to the widespread implementation of battery swapping stations. However, ensuring fairness in battery pricing remains a significant challenge since variations in battery health and performance among swapped batteries can result in user dissatisfaction and operational inefficiencies. This paper introduces a novel approach to enhance fairness in battery swapping by integrating a machine learning-based real-time prediction model with a pricing strategy based on remaining useful life (RUL) estimation to address this issue. The proposed solution comprises a real-time RUL estimation system and a dynamic pricing mechanism that ensures fair pricing based on battery health and performance. This integrated approach aims to improve user satisfaction and the operational efficiency of swapping stations. The paper evaluates various machine learning algorithms for real-time RUL estimation regarding accuracy, computation time, and memory usage. The results suggest that XGBoost provides the most suitable balance between accuracy and efficiency, making it an effective solution for real-world applications. Comparative analysis shows that the XGBoost model outperforms the second-best method (Random Forest) with a lower error (3.50 vs 3.79) while maintaining competitive computational efficiency (9.75 vs 8.52 seconds) and memory usage (2.12 vs 2.32 MB) when solving a typical numerical case study problem. The proposed approach has the potential to accelerate the adoption of electric vehicles and contribute to sustainability goals by promoting efficient battery utilization and fair pricing mechanisms.

This paper proposes solving the non-convex stochastic optimal power flow problem of a power system incorporating uncertain and intermittent renewable energy resources by an improved pelican optimization algorithm (IPOA). The POA, inspired from the foraging the behavior of pelicans, has stagnation problems and may trap into local optima. To avoid these, this paper has developed and implemented three novel improvements of mutation-based strategy, fitness distance balanced and exploitation-based gorilla troops strategies to enhance the exploitation and exploration strength of the traditional POA. The performance and effectiveness of the proposal are validated through statistical and non-parametric tests conducted via CEC 2019 test suite. In addition, IPOA is further used to solve a stochastic optimal power flow problem by integration of solar and wind energy to the modified IEEE 30-bus system to attain the lessen generation cost without and with inclusion of emissions. Statistical and non-parametric tests such as Wilcoxon ranking and Friedman tests validate the effectiveness of the proposed IPOA and the obtained least power generation costs and emissions for the considered numerical case studies.

Optimal energy hub scheduling (EHS) has emerged as a promising strategy for improving the efficiency and flexibility of power systems. Energy hubs (EHs) offer several advantages over conventional power grids, including enhanced flexibility, reduced emissions, and improved efficiency. However, EHS poses several challenges, including uncertainty, complexity, and computational burden. To tackle these challenges, this paper proposes an innovative optimal scheme for the operation of an integrated PV/wind energy system. The scheme incorporates a comprehensive set of components, including combined heat and power (CHP), power-to-gas (P2G), energy storage systems (ESSs), heat storage systems (HSSs), gas storage (GS), and electric boilers (EBs) and gas boilers (GBs). A demand response (DR) program is implemented for both electric and thermal loads to address the inherent uncertainty of renewable energy sources (RESs) and electrical load fluctuations. The proposed optimal management model is a multi-objective optimization problem aiming to minimize total losses, cost, and emissions while meeting energy demands. This novel approach offers significant advantages for utilities in terms of reducing losses, cost, and air pollution, contributing to a more sustainable energy system. The optimal management scheme is designed based on the optimized objective functions and implemented through steady-state energy analysis. Non-dominated sorting genetic algorithm III (NSGA-III) is employed to efficiently search for the optimal solutions. Scenario analysis is adopted to address the stochastic nature of RESs and load demand, and the Sim&Corrloss clustering strategy is used to reduce the computational burden. To demonstrate the effectiveness of the proposed approach, the results obtained from applying the proposed algorithm are compared with the results from analyzing the problem using GAMS software and the multi-objective seagull optimization algorithm (MOSOA). The proposed method enhances flexibility and ultimately increases system stability while maintaining diversity in energy sources. Additionally, the utilization of equipment such as various storage devices and P2G enhances system resilience, reducing load fluctuations and improving resource utilization. The results demonstrate that the proposed method significantly improves system performance and can effectively contribute to energy management in multi-energy systems. The superior performance of the proposed algorithm is demonstrated under various operating scenarios.