- Metallurgical Processes and Thermodynamics
- Metallurgy and Material Forming
- Aluminum Alloy Microstructure Properties
- Solidification and crystal growth phenomena
- High-Velocity Impact and Material Behavior
- High Temperature Alloys and Creep
- Fatigue and fracture mechanics
- Minerals Flotation and Separation Techniques
- Microstructure and Mechanical Properties of Steels
- Electromagnetic Launch and Propulsion Technology
- Intelligent Tutoring Systems and Adaptive Learning
- Machine Learning and Algorithms
- Nanofluid Flow and Heat Transfer
- AI-based Problem Solving and Planning
Indian Institute of Technology Madras
2023-2024
Tata Consultancy Services (India)
2015-2018
National University of Singapore
2014
Unified Mechanics Theory’s (UMT) entropy-based damage parameter, also known as the “Thermodynamic State Index” has been proven to be consistent and useful in predicting fatigue life of different metal alloys. In recent times, studies have demonstrated its applicability towards creep nickel-based superalloys under a limited set conditions. However, usefulness estimating at temperatures, loads for alloys not evaluated yet. this paper, INCONEL 600 alloy is modeled using Norton’s law modified...
In order to simulate and predict material's real-time responses for a component under complex mechanical thermal loads, continuum damage mechanics (CDM) is employed. However, majority of the models found in literature are phenomenological primarily based on curve fitting, which offer limited understanding underlying physics problem. A few physics-based have been developed that provide greater insights. Unified theory (UMT) one such approach captures entropy generation due various dissipative...
Random Forests are an effective ensemble method which is becoming increasingly popular, particularly for binary classification prediction problems.One of the most popular algorithms implementing Forest model Breiman and Cutler's algorithm this forms basis "randomForest" package in R.However, a implemented using has limitation, especially milieu limited computational power, that it cannot handle highly categorical data.In paper, we present one many techniques tried to improve performance...