Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 8, 2026Open Access

Deep Learning-Based Predictive Maintenance Framework Using CNN-LSTM Architecture for Industrial Rotating Machinery Fault Detection

View Full Paper
Ask AI
Bookmark
Share

Authors

KSK. Sujatha

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates enhanced fault detection in industrial machinery, suggesting cost savings in maintenance.

Key Points

  • The goal is to develop and validate a CNN-LSTM architecture for fault detection in industrial machinery.
  • Used time-series sensor data from a precision engineering facility over 18 months.
  • Implemented CNN for feature extraction from STFT spectrograms and LSTM for modeling temporal dependencies.
  • Tested the framework on 4,400 labeled samples across five fault classes.
  • Achieved 95.6% classification accuracy, 94.8% precision, and AUC of 0.981.
  • Outperformed standalone models: SVM (87.3%), Random Forest (91.2%), and LSTM (93.8%).
  • Projected 27% reduction in annual maintenance costs over three years.

Cite This Study

K. Sujatha (2025) studied this question.

synapsesocial.com/papers/69fd7f86bfa21ec5bbf080fehttps://doi.org/10.5281/zenodo.20052327
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Machine Learning-Based Predictive Fault Detection in Industrial Rotating Machinery Using Hybrid CNN-LSTM Architecture2026
  2. 2LSTM-CNN Hybrid Deep Learning Architecture for Predictive Maintenance and Remaining Useful Life Estimation in Rotating Industrial Machinery2026
  3. 3Machine Learning-Based Predictive Maintenance Framework for Rotating Equipment in Process Industries2026
  4. 4Machine Learning-Based Predictive Maintenance Framework for Rotating Equipment in Process Industries2026
  5. 5Research on Mechanical Fault Diagnosis and Prediction Technology Based on Deep Learning2024 · 1 citations