About My Research Projects

🔬 Physics-Informed Neural Networks (PINNs) & Deep Learning

1. Physics-Informed Reduced-Order Modeling for Lithium-Ion Battery State Estimation (Ongoing, 2026)

   • Collaborators: Dr. Jisha C.R. at the Advanced Mathematical Innovation Center for Artificial Intelligence and Scientific Computing (AMICAISC).

   • Overview: Currently developing reduced-order models that integrate physical principles with deep learning architectures to accurately estimate lithium-ion battery states.

2. Physics-Informed Neural Network Framework for Real-Time Battery Prognostics in EVs and Smart Grids (2025)

   • Collaborators: Nandana S., Vijithra A., and Dr. Jisha C.R. at the Advanced Mathematical Innovation Center for Artificial Intelligence and Scientific Computing (AMICAISC).

   • Overview: Designed a PINN framework aimed at real-time monitoring and predictive maintenance for electric vehicle (EV) batteries and smart grid energy systems.

3. Physics-Informed Deep Learning Framework for Bearing Degradation and Remaining Useful Life Prediction in Wind Turbines (2025)

   • Collaborators: Nandana S., Vijithra A., and Dr. Jisha C.R. at the Advanced Mathematical Innovation Center for Artificial Intelligence and Scientific Computing (AMICAISC).

   • Overview: Applied physics-informed methodologies to model bearing degradation and accurately forecast the remaining useful life of wind turbine components.

4. Physics-Informed Neural Network Framework for Engine Degradation Modeling (2025)

   • Collaborators: Nandana S., Vijithra A., and Dr. Jisha C.R. at the Advanced Mathematical Innovation Center for Artificial Intelligence and Scientific Computing (AMICAISC).

   • Overview: Modeled complex engine degradation processes by combining empirical data with fundamental physical laws using deep neural networks.

🧠 Explainable Artificial Intelligence (XAI) & Machine Learning

5. Explainable Machine Learning Frameworks for Predicting Hydrodynamic Residuary Resistance in Sailing Yachts (2026)

   • Advisor: Dr. Smrutiranjan Mohapatra, Assistant Professor, at Department of Mathematics, VSSUT, Burla.

   • Overview: Utilizing XAI techniques to build highly interpretable machine learning predictions for hydrodynamic resistance in marine engineering applications.

6. Benchmarking Interpretable Machine Learning Models for Heart Disease Risk Prediction using SHAP and TOPSIS (2025)

   • Advisor: Dr. Ambar Dutta, Associate Professor, at Amity University, Kolkata.

   • Overview: Evaluated and benchmarked various ML models for cardiovascular risk prediction, leveraging SHAP for feature interpretability and TOPSIS for multi-criteria decision-making.

7. Early-Stage Alzheimer's Disease Prediction Using Explainable Machine Learning (2025)

   • Advisor: Dr. Ambar Dutta, Associate Professor, at Amity University, Kolkata.

   • Overview: Developed predictive models for the early detection of Alzheimer's disease, prioritizing transparent and explainable algorithms suitable for medical diagnostics

🧮 Numerical Analysis & Computational Mathematics

8. Numerical Investigation of the Incompressible Navier-Stokes Equations (2025)

   • Advisor: Dr. Smrutiranjan Mohapatra, Assistant Professor, at Department of Mathematics, VSSUT, Burla.

   • Overview: Conducted in-depth numerical analysis and computational fluid dynamics modeling focused on solving incompressible Navier-Stokes equations.

9. Numerical Solutions of Ordinary Differential Equations (2024)

   • Advisor: Dr. Mansa Ranjan Sahoo, Associate Professor, at School of Mathematical Sciences, NISER, Bhubaneswar.

   • Overview: Explored advanced numerical techniques and algorithmic approaches for accurately solving complex ordinary differential equations.

10. A Modification of Newton Raphson Method (2024)

    • Advisor: Dr. Smrutiranjan Mohapatra, Assistant Professor, at Department of Mathematics, VSSUT, Burla.

    • Overview: Researched and developed modified mathematical approaches to the standard Newton-Raphson root-finding algorithm to improve computational efficiency and convergence rates.