Tamakloe Elvis is a graduating MPhil candidate in Computer Engineering at Kwame Nkrumah University of Science and Technology (KNUST) and a researcher with the Responsible Artificial Intelligence Lab (RAIL). His research investigates how artificial intelligence can drive sustainability, focusing on developing smart, data-driven solutions for critical infrastructure.
He has successfully defended his MPhil thesis, which pioneers a novel AI architecture for predictive maintenance in energy systems. His work specifically addresses the challenge of monitoring oil-immersed transformers, which are vital for a stable electricity supply. Anomalies in these transformers can lead to failures, causing power outages and economic disruption, yet accurately detecting these faults and predicting the equipment’s remaining lifespan remains a complex problem.
To tackle this, Elvis developed a novel Dynamic Multi-Scale Attention-based Convolutional Neural Network-Long Short-Term Memory (DMSA CNN-LSTM) architecture. His research demonstrates how fusing multi-modal data allows the model to dynamically select the most critical features across different time scales, capturing intricate details in transformer operations for superior anomaly detection, root cause localisation, and lifespan prediction. His proposed architecture shows a vast improvement over existing models, setting a new benchmark for deploying intelligent, preventative maintenance in real-world energy systems.
His research offers valuable insights into how scalable, data-driven AI can enhance the reliability and longevity of essential energy infrastructure, contributing to more resilient and sustainable smart grids.
He credits the lab’s guidance, mentorship, and collaborative environment for shaping his research journey. His work reflects RAIL’s mission of advancing responsible, Africa-centred AI that addresses real-world challenges and fosters sustainable development.