Showing papers by "A. S. Sekhar published in 2013"

Crack detection in a rotor by operational deflection shape and kurtosis using laser vibrometer measurements

Abstract: High speed rotating machinery operating with higher load capacity in extreme working conditions can induce fatigue crack into the systems. To avoid failure of the system, it has to be monitored continuously for cracks before it becomes critical. There is a need for the newer techniques that can detect even small cracks. The aim of the present work is to identify the crack in a rotor-bearing system using the concept of operational deflection shape and using kurtosis of vibration response. In the literature kurtosis is used for identifying bearing and gearbox defects. Higher kurtosis means more of the variance is due to infrequent extreme deviations, as opposed to frequent modestly sized deviations. In this study kurtosis deviation curve and amplitude deviation curves are also used for shaft crack detection. It has been noted that the kurtosis changes are significant when the crack is closer to the bearing locations. Single cracks and two cracks have been identified. The rotational laser vibrometer has been...

Application of Artificial Neural Networks for Identification of Unbalance and Looseness in Rotor Bearing Systems

Abstract: Rotating machinery is common in any industry. Rotating machinery in the modern era are designed for higher running speeds, tighter clearances and working under extreme conditions enhancing efficiency of the system to produce and transmit more power. All these lead to many rotordynamic challenges. Main cause of vibrations is faults in the rotating systems like unbalance, looseness, etc. In this paper a method is proposed to identify unbalance and looseness in rotor bearing system using artificial neural networks (ANN) by two different methods; one is by statistical features and the second by amplitude in frequency domain. In the first case statistical features are used to train and test the ANN, and in the second case amplitude in frequency domain is used to train and test the ANN. Experiments are conducted on the rotor bearing system running at 40 Hz and vibration data is collected by simulating different unbalance conditions in the rotor. And also experiments are conducted by creating looseness in the system by loosening the pedestal bolt. Various statistical features and amplitudes in frequency domain are extracted separately from this vibration data and are fed to neural network. It is observed that statistical features are giving good results over frequency domain amplitudes. ANNs are used to identify the unbalance severity and looseness. These results are useful for making maintenance decision.

Performance Analysis of a Rotating Labyrinth Seal With Radial Growth

Abstract: A computational methodology is proposed to predict the running clearance of a six-tooth straight-through rotating labyrinth seal numerically by taking into account both the centrifugal and thermal growths. Four different angular velocities ranging from 0 to 3000 rad/s are chosen to study the influence of rotation on the leakage flow rate. The detailed leakage flow fields and the structural deformations are presented. Further, different pressure ratios in the range of 1.1 to 2.5 have been investigated for a wide range of initial clearances. The methodology is validated against the available data in the literature. It is found out that there is a significant reduction in leakage flow rate by incorporating the radial growth for a particular operating condition. However, for a given initial clearance, the rotation has negligible effect on the reduction in the leakage flow rate, except at pressure ratios lower than 1.7. Further; the rotation has more prominent effect for smaller clearance values.Copyright © 2013 by ASME

Multi-objective optimisation of support characteristics of rotor bearing systems

Abstract: Rotors today are made highly flexible with ever increasing demand for higher speeds, and it is important to predict the rotor dynamic behaviour accurately. Supports play an important role and influence the magnitude of vibration in response to external forces. In the present study, frequency dependent polymeric support characteristics have been found by using a multi objective optimisation scheme for minimising the unbalanced response and maximising the stability limit speed. This is an improvement over the previous papers where a simple equally weighted sum of two objective functions was used. A novel hybrid of particle swarm optimisation (PSO) and augmented Lagrangian has been used here to find the global minima. Using appropriate constraints, the slope of the support characteristics versus frequency is restricted to realistic values. Rotor systems with rolling element bearings and plain cylindrical journal bearings are studied using this approach and higher stability limits are obtained.

Damage Identification and Quantification in Structures Using Wavelet Analysis

Abstract: In this study, displacement and strain energy mode shapes obtained from the structure, are processed using continuous spatial wavelet transforms for damage identification. A cantilever and simply supported beam are analysed in ANSYS to obtain mode shapes by performing modal analysis both for damaged and undamaged conditions. Different wavelets are processed to choose the best mother wavelet for detecting crack, a representation of damage. The dimensions of crack like depth are varied from 0.5 mm. Spatial distribution of wavelet coefficients obtained is used for identifying the damage. The quantification of the damage is done by using Lipchitz constant. It is observed that by using strain energy mode shapes the percentage of damage (crack) identification level can be improved than by using displacement mode shapes. The minimum damage level that the strain mode shape can identify is approximately 3% of the element thickness, whereas with displacement mode shapes, it is about 7% of the element thickness. Different types of crack configurations like single surface crack, double surface cracks and internal cracks are simulated and studied using both the modal methods. It is observed that the strain energy mode method could identify internal cracks also.