Projects
Title: Characterization of 400 volt high impedance fault with current and magnetic field measurements
This project characterizes high-impedance faults in low-voltage distribution networks using current and magnetic-field data and experimentally validates tree-branch fault signatures to enhance fault detection and system reliability.
Title: Cross-correlated scenario generation for renewable-rich power systems using implicit generative models
This project introduces a recurrent GAN-based method for generating realistic, cross-correlated scenarios in renewable-heavy power grids, improving long-term planning and reliability analysis by modeling variable dependencies without predefined distributions.
Title: Time-synchronized state estimation using graph neural networks in the presence of topology changes
This project develops a GNN-based state estimator for power systems that accommodates topology changes and sparse measurements, enhancing accuracy and robustness for real-time monitoring in dynamic grids.
Title: A facility for physical simulation of high impedance faults in low voltage networks
This project designs a physical simulation facility for high-impedance faults in low-voltage networks, enabling controlled testing of fault-detection technologies to improve safety and reliability in distribution systems.
Title: Feasibility of magnetic signature-based detection of low and high impedance faults in low-voltage distribution networks
This project examines magnetic-field-based fault detection in low-voltage grids, using simulations to compare with current measurements and to propose wavelet-based methods for improved fault identification.
Title: Data-driven flow and injection estimation in PMU-unobservable transmission systems
This project employs machine learning to estimate flows and injections in PMU-limited transmission networks, thereby providing efficient solutions to observability challenges in power system operations.
The integration of thermoelectric generators into improved cook stoves to convert waste heat into electricity, and proposes
This project explores the integration of thermoelectric generators into improved cook stoves to convert waste heat into electricity and proposes a prototype for sustainable energy in cooking applications.
Title: Classification of stages of a high impedance fault using sequential learning algorithms
This project employs sequential deep learning to classify high-impedance fault stages in distribution networks, enabling faster detection and response using sensor data.
Title: A review of optimum parameter values of a passive solar still and a design for southern Bangladesh
This project focuses on designing an affordable, portable passive solar still for clean water production in southern Bangladesh, reviewing optimal parameters like structure, condensers, reflectors, and wicks to maximize distillate yield, addressing water scarcity and heavy metal contamination in rural areas through renewable technology.
Title: Global solar radiation estimation from commonly available meteorological data for Bangladesh
This project develops and validates regression models to predict global solar radiation across Bangladesh using readily available meteorological data, enabling accurate assessment of solar potential for renewable energy planning without extensive instrumentation.
Title: Replacing diesel irrigation pumps with solar photovoltaic pumps for sustainable irrigation in Bangladesh: A feasibility study with HOMER
This project evaluates the economic and environmental feasibility of switching from diesel-powered to solar PV irrigation pumps in Bangladesh using HOMER simulations, thereby promoting sustainable agriculture by reducing emissions and costs through standalone systems suitable for rural applications.
Title: On the Effectiveness of Zero-Shot and Few-Shot Pretrained Language Models for Software Requirement Classification
This project evaluates zero- and few-shot language models for classifying functional and non-functional software requirements, demonstrating their effectiveness in reducing training needs and improving accuracy across datasets.
Title: Microcontroller-based 3-phase sequence indicator
This project designs a microcontroller system to indicate a 3-phase power sequence, ensuring proper rotation detection for machinery safety and improving electrical system efficiency.