Recent research:
The researcher’s work centers on proactive condition monitoring and fault detection in industrial systems, with a strong emphasis on machine learning, data-driven analytics, and scalable implementations. Across publications, the core themes include one-class learning for fault detection under limited labeled data, time-series and vibration signal analysis for health assessment, and practical architectures (including serverless and cloud-enabled solutions) to enable online or near-real-time monitoring. Methodologies combine unsupervised and semi-supervised learning (autoencoders, GANs, one-class SVM), frequency-domain analysis (FFT) of motor currents and vibration signals, and deployment perspectives that integrate agile design thinking with industry collaboration. The impact lies in demonstrating robust fault detection under data scarcity, illustrating how spectral features and learned representations correlate with fault severity, and proving scalable, industry-relevant architectures for continuous condition monitoring and early maintenance actions.

Key Topics:

• Artificial Intelligence, Machine Learning & Data Science
• Manufacturing Systems, Maintenance, Reliability & Predictive Maintenance (PHM)
• Sensors, Instrumentation & Measurement Systems