Open to opportunities

Haraprasad Badajena

@haraprasadbadajena

Message

Research scholar advancing ML-based fault diagnosis and predictive maintenance for electric motor health.

India

Message

What I'm looking for

I’m looking for a role where I can build ML + physics-driven fault diagnosis, digital-twin predictive maintenance, and real-time edge deployments—turning sensor data into reliable RUL and fault isolation for industrial motors, with room to publish and iterate fast.

I’m a Research Scholar in Fault Diagnostics & Prognostics at EE, IIT Kharagpur, where I designed and deployed multi-class fault detection, classification, and RUL prediction pipelines using CNN, GCN, and GraphSAGE, benchmarked against 8 state-of-the-art baselines. I also develop anomaly detection and fault isolation methods using physics-based, probabilistic, and machine learning approaches—then use digital twin simulations to optimize predictive maintenance and uncover degradation patterns.

Previously as a Junior Research Fellow at SRIC, IIT Kharagpur (sponsored by GAIL), I built and experimentally validated physics-based fault models for induction motors (BRB, bearing, ITSC, eccentricity, compound), achieving 90% accuracy on real test rigs. I’ve deployed real-time fault diagnosis pipelines with MCSA/ESA and advanced signal processing on edge platforms (Jetson Nano, Raspberry Pi) and implemented coupled-circuit digital twins in MATLAB for accurate physical-system replication, alongside supporting teaching as a Teaching Assistant.

Experience

Work history, roles, and key accomplishments

Current

Research Scholar

Current

IIT Kharagpur

May 2025 - Present (1 year 1 month)

Designed and deployed multi-class fault detection, classification, and RUL prediction pipelines for electric motor health monitoring, benchmarking against 8 state-of-the-art baselines. Integrated physics-based, probabilistic, and ML anomaly detection with a digital-twin workflow to support predictive maintenance optimization.

Fault Detection & Isolation (FDI)RUL Estimation CNN GCN GraphSAGE Anomaly Detection Digital Twins

Junior Research Fellow

SRIC, IIT Kharagpur

Aug 2022 - Jul 2023 (11 months)

Developed and experimentally validated physics-based fault models for induction motors (BRB, bearing, ITSC, eccentricity, compound), achieving 90% accuracy on real test rigs. Built real-time fault diagnosis pipelines using MCSA/ESA and advanced signal processing on embedded platforms, and created a coupled-circuit digital twin in MATLAB that replicated the physical system with 90% accuracy.

Physics Based Fault Modeling MCSA Signal Processing Induction Motor Fault Diagnosis Raspberry Pi