1. AI-Based Energy Modeling and Short-Term Load Forecasting for Renewable-Dominated Power Systems
This research focuses on developing an energy modeling framework using machine learning techniques (LSTM, GRU, or hybrid ML models) to forecast short-term load in power systems with high penetration of renewable energy. The work may compare traditional statistical methods with AI-based approaches and evaluate forecasting accuracy under variable weather and demand uncertainty.
2. IoT-Enabled Smart Renewable Energy Management with Adaptive Control Strategies
This study proposes an IoT-based monitoring and control architecture for distributed renewable energy resources such as solar PV and wind systems. Real-time sensor data are used to design adaptive control algorithms that optimize energy dispatch, reduce losses, and improve system reliability. Simulation combined with a small-scale hardware or virtual testbed can validate performance.
3. Machine Learning-Assisted Fault Detection and Classification in High-Voltage Transmission Systems
This topic investigates the use of machine learning and signal/image processing techniques for fault detection in high-voltage power systems. Time-frequency features from voltage and current signals (e.g., wavelet transforms or spectrograms) are used as inputs to ML classifiers. The goal is to improve detection speed and classification accuracy compared to conventional protection schemes.
4. Image Processing and Deep Learning for Solar PV Fault Diagnosis and Performance Degradation Analysis
This topic explores the use of image processing and deep learning techniques to detect faults in solar PV systems using thermal or visual images. Convolutional neural networks (CNNs) can be applied to identify hotspots, shading, and degradation patterns. The research emphasizes accuracy improvement and practical applicability for large-scale PV installations.