All Projects (0) Under Review (0) Published (0) Featured (0) Utility (0)

Projects

Graphical Abstract

Vision-Informed Multi-Modal Multi-Agent Reinforcement Learning for Active Distribution Network Scheduling

  • Introduced the first vision-aware framework for distribution-network scheduling by integrating image-derived soiling features into the cooperative control of solar plants, energy-storage systems, static-var compensators, and flexible water-pumping loads.
  • Built a CNN–ViT encoder for real-time generation loss estimation and fused it with grid measurements to enable a perception–control scheduling loop.
  • Implemented the architecture in PyTorch 2.6. The framework was validated using the DeepSolarEye dataset (44K annotated solar images) for soiling regression and applied to the realistic 2289-bus Nizwa grid in Oman, collected from Mazoon Electricity Company (MZEC).
  • Achieved R² = 0.91; reduced operational costs by 7.1%; lowered losses by 19.2%; decreased carbon emissions; and improved renewable-to-load alignment by 15%.
DOI
Graphical Abstract

Deep Learning-based Geospatial Mapping Framework for Large-Scale Electric Power Grids

  • Proposed a residual graph-convolutional-network framework capable of predicting geographic locations and connectivity of power-grid infrastructure at reduced computational cost.
  • Implemented the model in PyTorch with Torch Geometric 2.6.1. The study utilized the Mazoon Electricity Company (MZEC) dataset, comprising over 500K poles, 385K service points, and 23K substations across four governorates, as well as the Nigerian transmission grid dataset containing 56K components. All processed datasets were made publicly available via Zenodo.
  • Achieved testing accuracies of 95.88% (Oman) and 92.98% (Nigeria), along with near-perfect regression performance (R² = 0.9993 for Oman, 0.9960 for Nigeria). Training time was reduced by 50% compared to standard GCNs, with convergence achieved in 16–38 epochs versus 48–151 epochs for the baselines.
DOI
Graphical Abstract

AI-Driven Power Grid Path Prediction for Large-Scale Distribution Networks

  • Developed the first regression-based framework that formulates path prediction as a spatial-trajectory prediction problem.
  • Developed the model in TensorFlow 2.19.0 with Keras 3.10.0 that fuses sequential UTM coordinates with structured contextual features, including district, governorate, spatial clusters, voltage class, straightness, and cable length.
  • Conducted ablation and SHAP-based explainability analyses. Our results demonstrate that district and governorate embeddings are the dominant contributors, while voltage class, cable length, spatial cluster, and straightness offer negligible predictive value. The framework was validated on a large-scale dataset of 71K overhead conductor segments from the Mazoon Electricity Company (MZEC) grid in Oman.
See Under Review Publications See Featured Publications
Graphical Abstract

Neural Architecture Search via Reinforcement Learning for Image-based Solar Panel Soiling Loss Prediction

  • Proposed the first image-only regression framework for predicting soiling-induced power loss in solar panels using a reinforcement learning–driven neural architecture search (NAS) approach.
  • Evaluated the proposed NAS framework through two case studies: regression of soiling-induced power loss and multi-class classification of soiling severity levels.
  • Developed the framework in PyTorch 2.8.0+cu126 and utilized the DeepSolarEye dataset comprising 44K+ annotated solar-panel images for experimental validation.
See Under Review Publications See Featured Publications
Graphical Abstract

Robust Deep Learning Model for Spatiotemporal Forecasting of Renewable Energy Sources

  • Developed a spatiotemporal forecasting framework that fuses spatial feature extraction with temporal sequence learning for multi-horizon solar PV and wind power (WP) prediction.
  • Implemented the framework in TensorFlow 2.18.0 with Keras 3.8.0. All processed solar and wind datasets were made publicly available through Zenodo to support open research.
  • Achieved RMSE reductions of 3.8–25.7% for PV and 3.9–18.0% for WP (MAE reductions: 3.8–28.1% for PV and 4.2–32.6% for WP) relative to baselines, with strong robustness under 50% missing or noisy data.
DOI
Graphical Abstract

Microgrid Resource Planning Framework Using Multi-Agent Deep Reinforcement Learning Approach

  • Proposed a comprehensive microgrid resource planning framework that concurrently integrates EV charging scheduling and allocates solar farms, wind plants, battery energy storage systems, and EV charging stations.
  • Defined multi-dimensional objectives: minimizing investment–operational and emission costs while maximizing the voltage stability index and the average state of charge of EV fleets.
  • Developed the optimization framework in MATLAB R2021b, coupled with a multi-agent deep reinforcement learning–based EV charging scheduling model built in TensorFlow with Keras. Case studies were conducted on the modified IEEE 33-bus system, two practical 33-bus feeders in Bangladesh, and the Turkish 141-bus network using real-world meteorological and EV commuting data.
DOI
Graphical Abstract

Techno-Economic and Environmental Assessment of Electric Vehicle Charging Stations in Low-Inertia Microgrids

  • Proposed a comprehensive techno-economic and environmental assessment framework for microgrids integrating solar energy, wind energy, energy storage systems operated as virtual synchronous generators, and electric vehicle charging stations.
  • Defined three conflicting objectives: (i) Economic—minimizing the total net present cost and the levelized cost of electricity; (ii) Technical—reducing active power losses and frequency deviation and improving the voltage stability index; and (iii) Environmental—lowering carbon emissions.
  • Developed and implemented the optimization framework in MATLAB R2021b, and validated across two real-world case studies: the 51-bus Masirah Island network in Oman and the 141-bus Ankara distribution network in Turkey.
DOI
Graphical Abstract

Stochastic Multi-objective Architecture for Strategic Integration of Distributed Energy Resources in Microgrids

  • Proposed a comprehensive multi-objective framework for the optimal integration of battery energy storage systems (BESSs), photovoltaic (PV) systems, and wind turbines (WTs).
  • Defined the conflicting objectives as (i) minimizing the cumulative annual expenses of PV, WT, and BESS units, and (ii) maximizing the system-wide voltage stability margin. Uncertainties in renewable output and load demand were explicitly incorporated using probability distribution functions.
  • Developed and implemented the framework MATLAB R2021b and validated on the IEEE 33-bus, IEEE 69-bus, and Masirah Island (Oman) 51-bus distribution networks.
DOI
Graphical Abstract

Optimal Planning of Renewable Energy-Integrated Distribution System with Uncertainties

  • Proposed a novel optimization framework using MATLAB R2020a for the optimal placement and sizing of solar and wind plants.
  • Incorporated renewable-resource and demand uncertainty using Weibull distributions for wind speed, Beta distributions for solar irradiance, and normal distributions for load demand.
  • Validated the proposed approach on the IEEE 33-bus and IEEE 69-bus distribution networks, employing realistic solar irradiance, load, and wind speed datasets.
DOI
Graphical Abstract

Electric Vehicle Integration in Bangladeshi Residential Distribution Grid

DOI
Graphical Abstract

Optimal Load-Shedding in Distribution System

  • Proposed an advanced optimization technique for optimal load-shedding in renewable-integrated distribution networks subject to islanding conditions.
  • Defined the core objectives as maximizing the remaining load supplied after shedding, while simultaneously minimizing active power losses and improving voltage stability.
  • Implemented the optimization model in MATLAB R2020a and conducted extensive simulations on the IEEE 33-bus and IEEE 69-bus radial distribution networks across three distinct islanding scenarios.
DOI

Transient Stability Assessment of the Oman Grid under Large Penetration of Photovoltaic and Wind Energy Systems

  • Analyzed the stability challenges of Oman’s main interconnected system under high renewable penetration, aligned with the national development roadmap (500 MW Ibri II Solar IPP in 2021, 1,000 MW Manah Solar by 2025, 500 MW Ibri III by 2026, 500 MW A’Kamil Solar by 2027, and multiple wind projects totaling 600 MW between 2026–2027).
  • Proposed and validated virtual inertia systems integrated with battery energy storage, optimized via a swarm-based optimization algorithm in MATLAB R2021b and DIgSILENT PowerFactory.

Designing a Virtual Inertia Controller for the PDO Power System

  • Addressed low-inertia stability issues in the PDO grid under 400 MW (2026) and 600 MW (2030) renewable penetration by designing virtual inertia systems with battery energy storage.
  • Implemented the simulations in MATLAB R2021b and DIgSILENT PowerFactory, and optimized the size and placement of the inertia systems. The project provides a techno-economic roadmap for PDO’s 2026 and 2030 grid models.