The Diagnosis of Reciprocating Machinery by Bayesian Networks

A Bayesian Network is a reasoning tool based on probability theory and has many advantages that other reasoning tools do not have. This paper discusses the basic theory of Bayesian networks and studies the problems in constructing Bayesian networks. The paper also constructs a Bayesian diagnosis network of a reciprocating compressor. The example helps us to draw a conclusion that Bayesian diagnosis networks can diagnose reciprocating machinery effectively.

Adaptive sensitive frequency band selection for VMD to identify defective components of an axial piston pump

The underlying study investigates single valued neutrosophic entropy based adaptive sensitive frequency band selection for variational mode decomposition(VMD)for the purpose of identifying defective components in an axial pump.The proposed methodology is applied in the following steps.First,VMD is applied for decomposing vibration signals into various frequency bands,called as modes.After computing energy of each VMD,the lower(minimum)and upper(maximum)bounds from these energy readings are extracted for defect conditions,such as outer race,inner race,worn piston,faulty cylinder and valve plate,and blocked hole of the piston.Thereafter,energy interval ranges are obtained and further converted into the form of single valued neutrosophic sets(SVNSs).Then,the proposed neutrosophic entropy measure is deployed for quantifying the non-linear connection between each bearing defect conditions and various frequency bands.The mode having maximum neutrosophic entropy value is designated to the“most sensitive”frequency band.Thereafter,envelope demodulation is applied to the most sensitive selected frequency band for finding defective components.The proposed neutrosophic entropy and VMD based methodology is effective in providing a better insight for selecting suitable frequency band for carrying out envelope demodulation in comparison to existing methods.