Dynamic Analysis of Cracked Rotor in Viscous Medium and its Crack Diagnosis Using Intelligent Techniques

Abstract

Fatigue cracks have high potential to cause catastrophic failures in the rotor which can lead to catastrophic failure if undetected properly and in time. This fault may interrupt smooth, effective and efficient operation and performance of the machines. Thereby the importance of identification of crack in the rotor is not only for leading safe operation but also to prevent the loss of economy and lives.The condition monitoring of the engineering systems is attracted by the researchers and scientists very much to invent the automated fault diagnosis mechanism using the change in dynamic response before and after damage. When the rotor with transverse crack immersed in the viscous fluid, analysis of cracked rotor is difficult and complex. The analysis of cracked rotor partially submerged in the viscous fluid is widely used in various engineering systems such as long spinning shaft used drilling the seabed for the extracting the oil, high-speed turbine rotors, and analysis of centrifuges in a fluid medium. Therefore, dynamic analysis of cracked rotor partially submerged in the viscous medium have been presented in the current study. The theoretical analysis has been performed to measure the vibration signatures (Natural Frequencies and Amplitude) of multiple cracked mild steel rotor partially submerged in the viscous medium. The presence of the crack in rotor generates an additional flexibility. That is evaluated by strain energy release rate given by linear fracture mechanics. The additional flexibility alters the dynamic characteristics of cracked rotor in a viscous fluid. The local stiffness matrix has been calculated by the inverse of local dimensionless compliance matrix. The finite element analysis has been carried out to measure the vibration characteristics of cracked rotor partially submerged in the viscous medium using commercially available finite element software package ANSYS. It is observed from the current analysis, the various factors such as the viscosity of fluid, depth and position of the cracks affect the performance of the rotor and effectiveness of crack detection techniques. Various Artificial Intelligent (AI) techniques such as fuzzy logic, hybrid BPNN-RBFNN neural network, MANFIS and hybrid fuzzy-rule base controller based multiple faults diagnosis systems are developed using the dynamic response of rotating cracked rotor in a viscous medium to monitor the presence of crack. Experiments have been conducted to authenticate the performance and accuracy of proposed methods. Good agreement is observed between the results.