Title: AI, Innovation and Global Transformation: Interdisciplinary Perspectives on Technology, Business and Society
Editors: Dr. J. Preetha, and Dr. Siddhartha Mehrotra
ISBN: 978-81-69857-64-2
Chapter: 18
DOI: https://doi.org/10.59646/809/18
Authors: Dr. T. Albert, and Dr. P.K. Manikanda Pirapu
Abstract
Digital Twins (DT) bridge physical mechanical systems with high-fidelity virtual counterparts through bidirectional, real-time data synchronization. In mechanical engineering, DT frameworks modernize asset management by coupling multi-physics simulation, continuous condition monitoring, and prognostic analytics. This chapter presents an end-to-end cyber-physical architecture tailored for mission-critical mechanical systems, exemplified by high-speed industrial turbomachinery and rotary drivetrain assemblies. By combining reduced-order physics models with edge-acquired telemetry (triaxial vibration, thermography, acoustic emission), the architecture tracks degradation dynamics and predicts Remaining Useful Life (RUL). A comprehensive case study of an industrial centrifugal gas compressor drivetrain demonstrates automated anomaly detection, Bayesian-updated degradation tracking, and automated maintenance triggers. The experimental and analytical findings validate that integrating physical stress-life equations with machine learning yields higher prognostic accuracy, decreases unscheduled downtime by over 38%, and optimizes component longevity compared to conventional preventive strategies.
Keywords: Digital Twin, Predictive Maintenance, Physics-Informed Neural Networks, Remaining Useful Life, Condition Monitoring, Mechanical Degradation.