Comparison of decision tree-fuzzy and rough set-fuzzy methods for fault categorization of mono-block centrifugal pump

• Published in: Mechanical systems and signal processing Vol. 24; no. 6; pp. 1887 - 1906
• Main Authors: Sakthivel, N.R., Sugumaran, V., Nair, Binoy. B.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01-08-2010; Elsevier
• Subjects: Applied sciences; Artificial intelligence; C4.5 algorithm; Centrifugal pumps; Computer science; control theory; systems; Decision trees; Exact sciences and technology; Fault diagnosis; Fuzzy; Fuzzy logic; Hybrid systems; Industrial metrology. Testing; Learning and adaptive systems; Mechanical engineering. Machine design; Mechanical systems; Monitoring; Mono-block centrifugal pump; Principle component analysis; Pump and compressors (turbocompressors, fans, etc.); Rough set; Statistical features; Vibration; Mono-block centrifugal pump; Fault diagnosis; Fuzzy logic; Principle component analysis; C4.5 algorithm; Statistical features; Rough set; Measurement; Decision tree; Logic circuit; Fuzzy control; Rough set theory; Monitoring; Statistical analysis; Vibration; Probabilistic approach; Decision making; Momentum; Neural network; Decision analysis; Fuzzy set; Centrifugal pump; Fuzzy decision; Artificial intelligence; Fault diagnostic
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2010.01.008

Mono-block centrifugal pumps are widely used in a variety of applications. In many applications the role of mono-block centrifugal pump is critical and condition monitoring is essential. Vibration based continuous monitoring and analysis using machine learning approach is gaining momentum. Particularly, artificial neural networks, fuzzy logic have been employed for continuous monitoring and fault diagnosis. This paper presents the use of decision tree and rough sets to generate the rules from statistical features extracted from vibration signals under good and faulty conditions of a mono-block centrifugal pump. A fuzzy classifier is built using decision tree and rough set rules and tested using test data. The results obtained using decision tree rules and those obtained using rough set rules are compared. Finally, the accuracy of a principle component analysis based decision tree-fuzzy system is also evaluated. The study reveals that overall classification accuracy obtained by the decision tree-fuzzy hybrid system is to some extent better than the rough set-fuzzy hybrid system.