premkumarrs/motor-bearing-predictive-maintenance

How premkumarrs/motor-bearing-predictive-maintenance works

This repository is a motor-bearing predictive-maintenance system combining benchmark analysis, physical-rig sensing, ML diagnostics, and a Streamlit dashboard. The documented workflow runs from vibration and telemetry acquisition through windowed feature extraction and task-specific anomaly, fault, and RUL models to diagnostic results and visualization. Major branches include CWRU benchmark detection/classification, NASA C-MAPSS prognostics, and physical-rig analysis. Some links reflect README-described architecture; source excerpts are partial, so unshown implementation details are not inferred.

Main entry points: processed Evaluation Results · features.py Rig Features · anomaly_detection.py Rig Anomaly Detection · cwru_xgboost.py CWRU Fault Classifier

Components 18

Sensing and Data

ESP32 Motor Rig src/main.cpp
Rig Recordings data/raw
CWRU Dataset data/external/CWRU-dataset
NASA FD001 Data data/external/nasa_cmapps
Serial Reader dashboard.py

Signal Processing

Rig Features ml/fault_classification/features.py
CWRU Features ml/fault_classification/cwru_xgboost.py

ML Diagnostics

Rig Anomaly Detection ml/fault_classification/anomaly_detection.py
CWRU Anomaly Detection ml/fault_classification/cwru_anomaly_detection.py
CWRU Fault Classifier ml/fault_classification/cwru_xgboost.py
Grouped Validation ml/fault_classification/cwru_cross_validation.py
Health Scoring ml/fault_classification/health_score.py
Rig Classifier scripts/train_physical_rig_model.py
NASA RUL Model ml/rul_prediction/nasa_fd001_xgboost.py
Proxy RUL Model ml/rul_prediction/proxy_rul_lstm.py
Evaluation Results data/processed

Dashboard

Diagnostics Dashboard dashboard.py

Other components

Operator

Connections 22

ESP32 Motor Rig → Serial Reader: streams telemetry README.md
Serial Reader → Diagnostics Dashboard: feeds live view README.md
ESP32 Motor Rig → Rig Recordings: records samples
Rig Recordings → Rig Features: segments windows ml/fault_classification/features.py
Rig Features → Rig Anomaly Detection: supplies features ml/fault_classification/anomaly_detection.py
Rig Anomaly Detection → Health Scoring: provides anomaly results ml/fault_classification/health_score.py
Rig Features → Health Scoring: supplies features ml/fault_classification/health_score.py
Health Scoring → Proxy RUL Model: provides health data ml/rul_prediction/proxy_rul_lstm.py
Rig Features → Rig Classifier: supports classification
CWRU Dataset → CWRU Features: provides signals ml/fault_classification/cwru_xgboost.py
CWRU Features → CWRU Fault Classifier: fits classifier ml/fault_classification/cwru_xgboost.py
CWRU Dataset → CWRU Anomaly Detection: provides signals ml/fault_classification/cwru_anomaly_detection.py
CWRU Fault Classifier → Grouped Validation: shares model setup ml/fault_classification/cwru_cross_validation.py
NASA FD001 Data → NASA RUL Model: provides run data ml/rul_prediction/nasa_fd001_xgboost.py
Rig Anomaly Detection → Evaluation Results: writes evaluations ml/fault_classification/anomaly_detection.py
CWRU Anomaly Detection → Evaluation Results: writes evaluations ml/fault_classification/cwru_anomaly_detection.py
CWRU Fault Classifier → Evaluation Results: writes predictions ml/fault_classification/cwru_xgboost.py
Grouped Validation → Evaluation Results: writes validation ml/fault_classification/cwru_cross_validation.py
NASA RUL Model → Evaluation Results: writes predictions ml/rul_prediction/nasa_fd001_xgboost.py
Proxy RUL Model → Evaluation Results: writes results ml/rul_prediction/proxy_rul_lstm.py
Evaluation Results → Diagnostics Dashboard: visualizes evaluations README.md
Operator → Diagnostics Dashboard: views diagnostics README.md

Drawn by GitDiagram on November 9, 2026 from the repository's file tree, README and source files. Read as Markdown ·