Digital Twin Framework for Electric Motor Health Monitoring and Fault Prediction

The increasing adoption of Permanent Magnet Synchronous Motors (PMSMs) across industrial automation and electric mobility creates challenges for reliable, non-invasive health monitoring. This paper presents a prototype Digital Twin diagnostic screening system that integrates multi-phase current acquisition, automated signal windowing, 68-dimensional physics-informed feature extraction, ensemble machine learning classification, symmetrical sequence decomposition, and 3D virtual asset synchronization. National Instruments TDMS recordings captured at 5.0 kHz from physical PMSMs (1.0 kW, 1.5 kW, and 3.0 kW) are segmented into 256-sample windows. A 350-estimator Random Forest classifier processes a 68-channel representation comprising time-domain statistical moments, 6-band Fast Fourier Transform (FFT) sub-band energy densities, and inter-phase cross-correlations. Symmetrical sequence components (I0, I1, I2) are extracted to monitor negative-sequence imbalance. An interactive Three.js 3D WebGL Digital Twin provides real-time spatial fault localization, displaying a localized thermal hotspot beacon on Stator Slot #4 and dispatching automated ISO 10816 prescriptive maintenance work orders. Under rigorous recording-level, zero window-leakage grouped 5-fold cross-validation, the model achieves 82.29% CV accuracy and 89.12% holdout test accuracy with 100.0% precision and recall on intercoil faults.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24323749
Primary Topic
Machine Fault Diagnosis Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Digital Twin Framework for Electric Motor Health Monitoring and Fault Prediction

Vedant Kadam, Megha Sharma, Nathan Varghese, Ritessh Santra
6 links
Zenodo (CERN European Organization for Nuclear Research)
Machine Fault Diagnosis Techniques
preprint

Digital Twin Framework for Electric Motor Health Monitoring and Fault Prediction

Vedant Kadam, Megha Sharma, Nathan Varghese, Ritessh Santra
preprint en

Abstract

The increasing adoption of Permanent Magnet Synchronous Motors (PMSMs) across industrial automation and electric mobility creates challenges for reliable, non-invasive health monitoring. This paper presents a prototype Digital Twin diagnostic screening system that integrates multi-phase current acquisition, automated signal windowing, 68-dimensional physics-informed feature extraction, ensemble machine learning classification, symmetrical sequence decomposition, and 3D virtual asset synchronization. National Instruments TDMS recordings captured at 5.0 kHz from physical PMSMs (1.0 kW, 1.5 kW, and 3.0 kW) are segmented into 256-sample windows. A 350-estimator Random Forest classifier processes a 68-channel representation comprising time-domain statistical moments, 6-band Fast Fourier Transform (FFT) sub-band energy densities, and inter-phase cross-correlations. Symmetrical sequence components (I0, I1, I2) are extracted to monitor negative-sequence imbalance. An interactive Three.js 3D WebGL Digital Twin provides real-time spatial fault localization, displaying a localized thermal hotspot beacon on Stator Slot #4 and dispatching automated ISO 10816 prescriptive maintenance work orders. Under rigorous recording-level, zero window-leakage grouped 5-fold cross-validation, the model achieves 82.29% CV accuracy and 89.12% holdout test accuracy with 100.0% precision and recall on intercoil faults.

Zenodo (CERN European Organization for Nuclear Research)
Somaiya Vidyavihar University (IN)
Industry, innovation and infrastructure
Machine Fault Diagnosis Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon