Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models

Compressors and pumps are machines frequently used in petroleum and chemical industries for fluid transportation through flow systems to keep industrial processes running permanently. As their failure can produce costly disruption, developing fault detection and diagnosis tools is essential for accu...

Full description

Bibliographic Details
Main Authors: Diego Cabrera, Ruben Medina, Mariela Cerrada, René-Vinicio Sánchez, Edgar Estupiñan, Chuan Li
Format: Article
Language:English
Published: MDPI AG 2024-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/14/5/1710
_version_ 1797264840558903296
author Diego Cabrera
Ruben Medina
Mariela Cerrada
René-Vinicio Sánchez
Edgar Estupiñan
Chuan Li
author_facet Diego Cabrera
Ruben Medina
Mariela Cerrada
René-Vinicio Sánchez
Edgar Estupiñan
Chuan Li
author_sort Diego Cabrera
collection DOAJ
description Compressors and pumps are machines frequently used in petroleum and chemical industries for fluid transportation through flow systems to keep industrial processes running permanently. As their failure can produce costly disruption, developing fault detection and diagnosis tools is essential for accurately detecting and diagnosing faults. This research proposes a bi-dimensional representation of the vibration signal corresponding to the Mel Frequency Cepstral Coefficients (MFCC) and their first two derivatives as features. The pseudo-periodic nature of the fault signature in rotating machines is exploited to put forward an efficient and accurate patch-wise fault classification method. This approach enables the classification of 13 combined types of faults in a multi-stage centrifugal pump and 17 faults in a reciprocating compressor. Classification is performed using the Long Short-Term Memory (LSTM) network, the bidirectional Long Short-Term Memory (BiLSTM) neural network, and the Convolutional Neural Network (CNN). Accurate classification over 99% is attained, showing that the proposed feature extraction procedure correctly classifies a large set of faults simultaneously appearing in such rotating machines.
first_indexed 2024-04-25T00:35:18Z
format Article
id doaj.art-465820a184b944e1b83ee41259cf38b5
institution Directory Open Access Journal
issn 2076-3417
language English
last_indexed 2024-04-25T00:35:18Z
publishDate 2024-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj.art-465820a184b944e1b83ee41259cf38b52024-03-12T16:38:35ZengMDPI AGApplied Sciences2076-34172024-02-01145171010.3390/app14051710Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning ModelsDiego Cabrera0Ruben Medina1Mariela Cerrada2René-Vinicio Sánchez3Edgar Estupiñan4Chuan Li5GIDTEC, Universidad Politécnica Salesiana, Cuenca 010105, EcuadorCIBYTEL—Engineering School, Universidad de Los Andes, Mérida 5101, VenezuelaGIDTEC, Universidad Politécnica Salesiana, Cuenca 010105, EcuadorGIDTEC, Universidad Politécnica Salesiana, Cuenca 010105, EcuadorMechanical Engineering Department, Universidad de Tarapacá, Arica 1010069, ChileSchool of Mechanical Engineering, Dongguan University of Technology, Dongguan 523808, ChinaCompressors and pumps are machines frequently used in petroleum and chemical industries for fluid transportation through flow systems to keep industrial processes running permanently. As their failure can produce costly disruption, developing fault detection and diagnosis tools is essential for accurately detecting and diagnosing faults. This research proposes a bi-dimensional representation of the vibration signal corresponding to the Mel Frequency Cepstral Coefficients (MFCC) and their first two derivatives as features. The pseudo-periodic nature of the fault signature in rotating machines is exploited to put forward an efficient and accurate patch-wise fault classification method. This approach enables the classification of 13 combined types of faults in a multi-stage centrifugal pump and 17 faults in a reciprocating compressor. Classification is performed using the Long Short-Term Memory (LSTM) network, the bidirectional Long Short-Term Memory (BiLSTM) neural network, and the Convolutional Neural Network (CNN). Accurate classification over 99% is attained, showing that the proposed feature extraction procedure correctly classifies a large set of faults simultaneously appearing in such rotating machines.https://www.mdpi.com/2076-3417/14/5/1710fault diagnosismel frequency cepstral coefficientsconvolutional neural networkslong short term memoryreciprocating compressorsmulti-stage centrifugal pumps
spellingShingle Diego Cabrera
Ruben Medina
Mariela Cerrada
René-Vinicio Sánchez
Edgar Estupiñan
Chuan Li
Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
Applied Sciences
fault diagnosis
mel frequency cepstral coefficients
convolutional neural networks
long short term memory
reciprocating compressors
multi-stage centrifugal pumps
title Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
title_full Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
title_fullStr Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
title_full_unstemmed Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
title_short Improved Mel Frequency Cepstral Coefficients for Compressors and Pumps Fault Diagnosis with Deep Learning Models
title_sort improved mel frequency cepstral coefficients for compressors and pumps fault diagnosis with deep learning models
topic fault diagnosis
mel frequency cepstral coefficients
convolutional neural networks
long short term memory
reciprocating compressors
multi-stage centrifugal pumps
url https://www.mdpi.com/2076-3417/14/5/1710
work_keys_str_mv AT diegocabrera improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels
AT rubenmedina improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels
AT marielacerrada improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels
AT reneviniciosanchez improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels
AT edgarestupinan improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels
AT chuanli improvedmelfrequencycepstralcoefficientsforcompressorsandpumpsfaultdiagnosiswithdeeplearningmodels