Fault Diagnosis of Air Compressor (AC) System using Local Mean Decomposition (LMD) and Logistic Regression (LR) Machine Learning Classifier

Year : 2026 | Volume : 14 | Special Issue 01 | Page : 416 427
    By

    Atul Dhakar,

  • Bhagat Singh,

  • Pankaj Gupta,

  1. Assistant Professor, Department of Mechanical Engineering, University Institute of Technology (UIT), RGPV, Shivpuri, Madhya Pradesh, India
  2. Head & Associate Professor, Department of Mechanical Engineering, Jaypee University of Engineering and Technology, Guna, Madhya Pradesh, India
  3. Assistant Professor, Department of Mechanical Engineering, Jaypee University of Engineering and Technology, Guna, Madhya Pradesh, India

Abstract

This article presents a detailed and systematic procedure for performing fault diagnosis in an air compressor (AC) system by analyzing the audio signals generated during its operation. The analysis covers both normal (healthy) conditions and seven distinct types of faults, including bearing failure, flywheel malfunction, inlet valve leakage, outlet valve leakage, non-return valve failure, piston ring defect, and rider belt issues. To acquire the acoustic signals, the researchers utilized a unidirectional microphone in combination with National Instruments hardware, specifically the NI 9234 data acquisition device and the NI 9172 USB interface. These tools ensured accurate and high-quality audio signal collection from the AC system under various operating conditions. Once the audio data were collected, they were subjected to a sophisticated, non-traditional signal processing method known as Local Mean Decomposition (LMD). This technique is especially effective for handling non-linear and non-stationary signals commonly found in mechanical systems. Following decomposition, six key statistical indicators (SIs) were computed to extract distinguishing fault features. These indicators included the mean of signals (MS), variance (σ²VS), mean of the root square (MRS), mean of the root amplitude (MRA), mean of the absolute amplitude (MAA), and the kurtosis index (KU). The extracted features were classified using a Logistic Regression (LR) machine learning algorithm. The integrated use of LMD, the kurtosis index, and the LR classifier led to a classification accuracy of 90.30%, demonstrating strong effectiveness in identifying faults within the AC system.

Keywords: AC system, LMD, statistical indicators, logistic regression classifier, fault diagnosis.

[This article belongs to Special Issue under section in Journal of Polymer & Composites (jopc)]

How to cite this article:
Atul Dhakar, Bhagat Singh, Pankaj Gupta. Fault Diagnosis of Air Compressor (AC) System using Local Mean Decomposition (LMD) and Logistic Regression (LR) Machine Learning Classifier. Journal of Polymer & Composites. 2026; 14(01):416-427.
How to cite this URL:
Atul Dhakar, Bhagat Singh, Pankaj Gupta. Fault Diagnosis of Air Compressor (AC) System using Local Mean Decomposition (LMD) and Logistic Regression (LR) Machine Learning Classifier. Journal of Polymer & Composites. 2026; 14(01):416-427. Available from: https://journals.stmjournals.com/jopc/article=2026/view=237001


References

  1. Ruonan L, Boyuan Y, Enrico Z, Xuefeng C. Artificial intelligence for fault diagnosis of rotating machinery: A review. Mech Syst Signal Process. 2018; 108:33–47.
  2. Blodt M, Grabjon P, Raison B, Rostaing G. Models for bearing damage detection in induction motors using stator current monitoring. IEEE Trans Ind Electron. 2008;55(4):1813–1822.
  3. Haji M, Toliyat HA. Pattern recognition – A technique for induction machines rotor broken bar detection. IEEE Trans Energy Convers. 2001; 16(4):312–317.
  4. Nandi S, Toliyat HA, Xiaodong L. Condition monitoring and fault diagnosis of electrical motors – A review. IEEE Trans Energy Convers. 2005; 20(4):719–729.
  5. Yan R, Gao R. Hilbert-Huang transform-based vibration signal analysis for machine health monitoring. IEEE Trans Instrum Meas. 2006; 55(6):2320–2329.
  6. Scanlon P, Kavanagh DF, Boland FM. Residual life prediction of rotating machines using acoustic noise signals. IEEE Trans Instrum Meas. 2013; 62(1):95–108.
  7. McInerny SA, Dai Y. Basic vibration signal processing for bearing fault detection. IEEE Trans Educ. 2003; 46(1):149–156.
  8. Li R, He D. Rotational health machine monitoring and fault detection using EMD-based acoustic emission feature quantification. IEEE Trans Instrum Meas. 2012; 61(4):990–1001.
  9. Verma NK, Salour AL. Intelligent Condition Based Monitoring. Springer. 2020; 256.
  10. Verma NK, Sevakula RK, Dixit S, Salour AL. Intelligent condition-based monitoring using acoustic signals for air compressors. IEEE Trans Reliab. 2015; 99:1–19.
  11. Delio T, Tlusty J, Smith S. Use of audio signals for chatter detection and control. J Manuf Sci Eng Trans ASME. 1992; 114(2):146–157.
  12. Verma NK, Kadambari J, Abhijit B, Tanu S, Salour AL. Finding sensitive sensor positions under faulty condition of reciprocating air compressors. IEEE Recent Adv Intell Comput Syst (RAICS). 2011: 242–246.
  13. Shrivastava Y, Singh B, Sharma A. Identification of chatter in turning operation using WD and EMD. Mater Today Proc. 2018; 5(11):23917–23926.
  14. Jun W, Guifu D, Zhongkui Z, Changquing S, Qingbo H. Fault diagnosis of rotating machines based on the EMD manifold. Mech Syst Signal Process. 2020; 135:1–21.
  15. Yongbo L, Shubin S, Zhiliang L, Xihui L. Review of local mean decomposition and its application in fault diagnosis of rotating machinery. J Syst Eng Electron. 2019; 30(4):799–814.
  16. Estrada A. Optimized LMD method and its applications in rolling bearing fault diagnosis. Bioorg Med Chem Lett. 2012; 23(6):149–152.
  17. Maurya S, Singh V, Dhar NK, Verma NK. Improved EMD local energy with SVM for fault diagnosis in air compressor. Adv Intell Syst Comput. 2019; 799:81–92.
  18. Yaguo L. Fault diagnosis of rotating machinery based on empirical mode decomposition. Smart Sens Meas Instrum. 2017; 26:259–292.
  19. Peng ZK, Tse PW, Chu FL. A comparison study of improved Hilbert-Huang transform, and wavelet transform: Application to fault diagnosis for rolling bearing. Mech Syst Signal Process. 2005; 19(5):974–988.
  20. Feng Z, Liang M, Chu F. Recent advances in time-frequency analysis methods for machinery fault diagnosis: A review with application examples. Mech Syst Signal Process. 2013; 38(1):165–205.
  21. Gupta P, Singh B. Local mean decomposition and artificial neural network approach to mitigate tool chatter and improve material removal rate in turning operation. Appl Soft Comput J. 2020;96.
  22. Bin GF. Early fault diagnosis of rotating machinery based on wavelet packets – Empirical Mode Decomposition feature extraction and neural network. Mech Syst Signal Process. 2012; 27(1):696–711.
  23. Pandya DH, Upadhyay SH, Harsha SP. Fault diagnosis of rolling element bearing by using multinomial logistic regression and wavelet packet transform. Soft Comput. 2014; 18:255–266.
  24. Yan J, Lee J. Degradation assessment and fault modes classification using logistic regression. J Manuf Sci Eng. 2004; 124(4):912–914.
Special Issue Subscription Review Article
Volume 14
Special Issue 01
Received 13/09/2025
Accepted 29/09/2025
Published 14/02/2026
Publication Time 154 Days
79

Login


My IP

PlumX Metrics