Our strong R&D background enables us to harness the latest AI techniques and dig deep into complex subject matter to deliver the most optimized solutions, even in challenging situations such as limited data availability. We develop deep learning and hybrid models for a diverse range of applications that include demand forecasting, cybersecurity in smart infrastructure, predictive maintenance with minimal sensors, energy efficiency and management, and health diagnostics. Additionally, we explore the applications of the latest versions of the fast evolving AI technologies in text recognition, gesture control, healthcare, and customer support—creating practical, reliable, and cost effective solutions that improve safety, performance, and user experience across diverse sectors and infrastructures.
AI & Cybersecurity
Our research explores advanced AI-driven cybersecurity solutions for smart infrastructure including smart power systems. We develop deep learning and hybrid models to detect cyberattacks, mitigate risks, and enhance system resilience—especially in the context of increasing renewable integration and electricity theft.
Research Publications
- Mitigating Cyber Risks in Smart Cyber-Physical Power Systems Through Deep Learning and Hybrid Security Models, IEEE Access, DOI: 10.1109/ACCESS.2025.3545637
- Deep Learning-based Cyber Attacks Detection in Power Grids with Increasing Renewable Penetration, IEEE World AI IoT Congress (AIIoT), Seattle, USA, May 2024
- A Biphasic Machine Learning Approach for Detecting Electricity Theft Cyberattacks in Smart Grids, in review IEEE Trans. Smart Grids
AI & Predictive Maintenance
Our AI and Predictive Maintenance (PdM) research focuses on developing data-driven and neural network-based solutions for early fault detection and health monitoring in industrial systems. From multi-motor machines to HVAC and battery storage systems, we enable accurate anomaly detection with minimal sensors, improving reliability, safety, and operational efficiency across critical infrastructures.
Research Publications
- Automatic Control and Health Monitoring of a 3-Dimensional Overhead Crane with Minimally Required Sensor Devices, 21st International Conference on Informatics in Control, Automation and Robotics (ICINCO 2024), Porto, Portugal, Nov. 18-20, 2024
- A Data-Driven Approach Based on Artificial Neural Networks for the Detection and Classification of Bearing Anomalies in Power Generation Plants, IEEE World AI IoT Congress (AIIoT), Seattle, USA, 7-10 June 2023
- Neural Network-Based Pipeline for Detection of Sensor Anomalies in Battery Energy Storage Systems, International Conference on Industrial and Information Systems (ICIIS’2023), Sep 2023, Kandy
AI & Battery Mgmt
Our AI research in battery management systems focuses on accurate degradation assessment and Remaining Useful Life (RUL) prediction for lithium-ion batteries. Using deep learning and simulated impedance spectroscopy, we enhance battery health diagnostics and lifecycle optimization—supporting smarter, safer, and more reliable energy storage solutions.
Research Publications
- EV Battery Degradation Assessment Under Standard Drive Cycle Using Simulated EIS, Vehicles 2025, 7(1), 21; https://doi.org/10.3390/vehicles7010021
- Review on Li-Ion Battery Parameter Extraction Methods, IEEE Access, Vol. 11, 2023
- Accurate Prediction of Remaining Useful Life for Lithium-ion Battery Cells Using Deep Neural Networks, IEEE World AI IoT Congress (AIIoT), Seattle, USA, May 2024
- Assessment of Lithium-Ion Battery Degradation and its impact on ECM using Simulated Impedance Spectroscopy, ECCE Europe 2024, Darmstadt, Germany, Sep 2024
AI & Energy Efficiency/Mgmt
Our AI research in energy efficiency and management focuses on deep learning and adaptive algorithms for optimizing energy use in smart residential and commercial buildings. We develop advanced Non-Intrusive Load Monitoring (NILM), HVAC optimization, and load classification techniques to enable intelligent resource and operationally efficient infrastructures.
Research Publications
- A Self-Adaptive Deep Learning Framework for Non-Intrusive Load Monitoring: Addressing Aging Appliance Challenges with Transfer Learning and Pseudo Labeling, IEEE Access
- NILM for Commercial Buildings: Deep Neural Networks Tackling Non-Linear and Multi-Phase Loads, Energies (Section F: electrical engineering), Vol. 17, Issue 15, Aug. 2024
- Maximizing Efficiency in Commercial Power Systems with an Optimized Load Classification and Identification Method Using Deep Learning and Ensemble Techniques, IEEE World AI IoT Congress (AIIoT), Seattle, USA, 7-10 June 2023
- Deep Learning-based Non-Intrusive Load Monitoring for Three-Phase Systems,” IEEE Access, Vol. 11, 2023
- A Stepwise Approach-Based Optimal Energy Utilization Scheme for HVAC Secondary Chilled Water Pumps in Commercial Buildings,” International Energy Journal 22, pp 389-400, Dec 2022
- Self-Adaptive Non-Intrusive Load Monitoring Using Deep Learning, IEEE World AI IoT Congress (AIIoT), Seattle, USA, May 2024
- Indoor and Outdoor Conditions Utilized Energy Saving Scheme for HVAC Cooling Water Systems in Smart Commercial Buildings, IEEE Canada Electrical Power and Energy Conference (EPEC), Oct., 2021
An Optimal Electrical Energy Management Scheme for Future Smart Homes, IEEE 8th International Conference on Smart Grid Engineering, Oshawa, Canada, Aug., 2020
Miscellaneous
Our diverse research spans text recognition, gesture-based smart home control, remote medical systems, solar energy optimization, demand forecasting, and automated customer support.
Research Publications
- A Multilingual Semantic Fusion Network for Text Recognition in the Wild Authors, Journal of Electronic Imaging (JEI)
- Techno-economic Feasibility Assessment of Implementation of Solar Tracking for Utility Scale Solar Photovoltaic Plants in Sri Lanka, IEEE GTD Asia 2019, Bangkok, March, 2019
- A Portable Remote Medical Consultation System for the Use of Distant Rural Communities, Int Conf on Materials, Electronics, and Information Engineering (ICMEIE’15), Rajshahi, B’Desh, June, 2015
- Vision-based Hand Gesture Recognition System for Appliance Control in Smart Homes, IEEE Int Conf Signal Processing, Communications, and Computing, Hong Kong, Aug., 2012
- A Short-term Electricity Demand Forecasting Method for Smart Meters, IEEE Int Conf Information and Automation for Sustainability, Beijing, China, Aug., 2012
Automated Question Answering for Customer Helpdesk Applications, 6th IEEE Int. Conf. Industrial and Information Systems, Aug., 2011
