Bayesian approach and time series dimensionality reduction to LSTM-based model-building for fault diagnosis of a reciprocating compressor

Engineering & Materials Science

• Reciprocating compressors 99%
• Long short-term memory 84%
• Dimensionality reduction 78%
• Time series 62%
• Failure analysis 62%
• Machinery 52%
• Deep learning 32%
• Compressed air 21%
• Compressors 15%
• Learning algorithms 15%
• Feature extraction 13%
• Industry 7%

Medicine & Life Sciences

• Deep Learning 100%
• Bayes Theorem 88%
• Long-Term Memory 83%
• Short-Term Memory 68%
• Compressed Air 54%
• Vibration 40%
• Industry 33%
• Psychology Recognition 32%