Skip to content

lifesailor/data-driven-predictive-maintenance

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

25 Commits
 
 
 
 

Repository files navigation

Predictive maintenance

A curated list of predictive maintenance resources. Inspired by awesome-ml and awesome-anomaly-detection.



What is predictive maintenance?

Predictive maintenance(PdM) techniques are designed to help determine the condition of in-service equipment in order to predict when maintenance should be performed.

The main promise of predictive maintenance is to allow convenient scheduling of corrective maintenance and to prevent unexpected equipment failures. The key is "the right information in the right time".

Prognostics, Fault detection is similar concept to predictive maintenance.



Table of Content

  1. Concept of predictive maintenance

  2. Predictive maintenance depending on the type of data.

    1. Vibration
    2. Sound
    3. Process Parameter
  3. Predictive maintenance technique

    1. Signal Processing
    2. Machine Learning
    • Classification
    • Anomaly Detection
  4. Predictiev maintenance conference



1. Concept of predictive maintenance

Publication

  • An introduction to predictive maintenance | R.Keith Mobley | [Elsevier Science(USA)' 2002] | pdf

    • This book clearly explains key concepts and basic techniques of predictive maintenance. At the first part of the book, it is explained that why predictive maintenance is important and how to justify it in financial aspects. The second part presents various techinques for each data type. Vibartion analysis, thermography, tribology, process parameters, ultrasonics, visual inspection, operating dynamics analysis are described in each chapter.
  • A Survey of Data-Driven Prognostics | Mark A. Schwabacher |[NASA Ames Research Center' 2006] | pdf


Slide

  • Introduction to Prognostics | N.Scott Clements | [Annual Conference of the PHM Society' 2011] | pdf
  • Introduction to Prognostics | Kai Goebel, Abhinav Saxena, Matt Daigle, Jose Celaya, Indranil Roychoudhury, Scott Clements |[NASA Prognostics CoE' 2012] | pdf
    • The above slides explain prognostics is estimation of the remaining useful life of a component. Remaining useful life of a component can be expected to continue operating within its given specifications. The brief explanation of three methods for prognostics are provided.

Essay

  • (ko) 2017 PHMAP Conference Review | Seulgi Lee | [Data Mining & Quality Analytics Lab] | html
    • I lack mechanically engineering and physical background. Such knowledge, of course, may be necessary, but the important thing is to properly predict the failure of the equipment. This resource gave me the courage to approach the problem as a data scientist, not as a mechanical engineer.



2. Predictive maintenance depending on the type of data

1. vibration

Slide

  • Beginning vibration analysis with basic fundamentals | Jack Peters | [CTC' 2015] | pdf

Video

  • Vibration Analysis Know-How: Quick Intro to Vibration Analysis | Jason Tranter | [LUDECAINC' 2015] | Youtube
    • Provides good introduction to vibration analysis

Publication

  • Vibration Analysis and Diagnostic Guide | Jaafar Alsalaet | [Unversity of Barash' 2012] | pdf
    • Provides good introduction of traditional vibration analysis in English.
  • (ko) Vibration and Equipment Handbook | Byungjoon, Lee | [Korea Electronic Power Corporation' 1998] | pdf
    • Provides good explanation of traditional vibration analysis in Korean.
  • (book) Structural health monitoring: a machine learning perspective |Charles R. Farrar, Keith Worden| [John Wiley & Sons, Ltd' 2013]| pdf
    • Machine learning for vibration analysis.
  • Deep Learning and Its Applications to Machine Health Monitoring: A Survey | Rui Zhao, Ruqiang Yan, Zhenghua Chen, Kezhi Mao, Peng Wang, and Robert X. Gao | [IEEE Transactions on Neural Networks and Learning Systems' 2016] |pdf
    • Deep learning for vibartion analysis.

2. sound

Interview

  • Acoustic signals for predictive maintenance: An interview with Amnon Shenfeld of 3DSignals | Robin Whitlock | [Renewable Energy Magazine' 2017] | html
    • The interview explains the advantage of using sound data for predictive maintenance. Only few sound sensors or microphones you can pick up acoustic behavior of a whole bunch of machines, even track complex processes.

Publication

  • Sound based fault detection system | Teemu Tossavainen | [Aalto university' 2015] | pdf

  • Residual Error Based Anomaly Detection Using Auto-Encoder in SMD Machine Sound | Dong Yul Oh, Il Dong Yun | [sensors' 2018] | pdf

    • Transform the sound into a spectrogram and applied convolutional autoencoder to detect anomalies.
  • Anomaly Detection Technique in Sound to Detect Faulty Equipment| Hisashi Uematsu, Yuma Koizumi, Shoichiro Saito, Akira Nakagawa, and Noboru Harada| [NTT Technical review' 2017] | pdf

    • Explains the reason why it is hard to apply supervised machine learning techinque for predictive maintenance using sound data.
  • Unsupervised Detection of Anomalous Sound based on Deep Learning and the Neyman-Pearson Lemma | Yuma Koizumi, Shoichiro Sait, Hisashi Uematsu, Yuta Kawachi, and Noboru Harada | [arXiv' 2018] | pdf


3. process parameter

Publication

  • Recurrent auto-encoder model for large-scale industrial sensor signal analysis|Timothy Wong and Zhiyuan Luo| Royal Holloway | [Arxiv' 2018] | pdf


3. Predictive maintenance techinque

  1. Signal Processing
    • Understand the Fourier transform and its applications | Mike X Cohen | Udemy
    • Signal Processing problems solved in MATLAB and in python | Mike X Cohen | Udemy
  2. Machine Learning
    1. Classification
    2. Anomaly Detection


4. Predictive maintenance conference

  • PHM Society | html
  • (ko) Korean Society for Prognostics and Health Managment | html

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published