A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Evaluation of performance of vibration signatures for condition monitoring of worm gearbox by using ANN
https://orcid.org/http://orcid.org/0000-0002-3343-8688
The worm wheel is the critical element and is vulnerable to failure due to fault occurring on it that leads to downtime hampering productivity. The condition monitoring can predict deteriorating health due to the fault. This work presents the experimental investigation of the worm gearbox carried out by the design of experiments (DOE) that utilizes amplitudes of denoised vibration signatures. During the experiments, trials are designed by the response surface method Box-Behnken DOE method. The cases considered for a single fault are (a) healthy gearbox (b) defective bearing inner race (c) defective outer race (d) defective worm wheel followed by cases comprising of a combination of two faults (e) fault on bearing inner and outer race (f) faulty worm wheel and bearing inner race (g) defective worm wheel and bearing outer race (h) fault on all three was acquired. The statistical parameters extracted from the acquired vibration signatures were used as input to train the ANN model and the performance is evaluated. The results show that the worm wheel is predominant for the fault over the other entities. ANN model predicts fault with an accuracy of 92.9%. Research outcomes envisage that the methodology used has good solidity to improve the performance.
Evaluation of performance of vibration signatures for condition monitoring of worm gearbox by using ANN
https://orcid.org/http://orcid.org/0000-0002-3343-8688
The worm wheel is the critical element and is vulnerable to failure due to fault occurring on it that leads to downtime hampering productivity. The condition monitoring can predict deteriorating health due to the fault. This work presents the experimental investigation of the worm gearbox carried out by the design of experiments (DOE) that utilizes amplitudes of denoised vibration signatures. During the experiments, trials are designed by the response surface method Box-Behnken DOE method. The cases considered for a single fault are (a) healthy gearbox (b) defective bearing inner race (c) defective outer race (d) defective worm wheel followed by cases comprising of a combination of two faults (e) fault on bearing inner and outer race (f) faulty worm wheel and bearing inner race (g) defective worm wheel and bearing outer race (h) fault on all three was acquired. The statistical parameters extracted from the acquired vibration signatures were used as input to train the ANN model and the performance is evaluated. The results show that the worm wheel is predominant for the fault over the other entities. ANN model predicts fault with an accuracy of 92.9%. Research outcomes envisage that the methodology used has good solidity to improve the performance.
Evaluation of performance of vibration signatures for condition monitoring of worm gearbox by using ANN
https://orcid.org/http://orcid.org/0000-0002-3343-8688
The worm wheel is the critical element and is vulnerable to failure due to fault occurring on it that leads to downtime hampering productivity. The condition monitoring can predict deteriorating health due to the fault. This work presents the experimental investigation of the worm gearbox carried out by the design of experiments (DOE) that utilizes amplitudes of denoised vibration signatures. During the experiments, trials are designed by the response surface method Box-Behnken DOE method. The cases considered for a single fault are (a) healthy gearbox (b) defective bearing inner race (c) defective outer race (d) defective worm wheel followed by cases comprising of a combination of two faults (e) fault on bearing inner and outer race (f) faulty worm wheel and bearing inner race (g) defective worm wheel and bearing outer race (h) fault on all three was acquired. The statistical parameters extracted from the acquired vibration signatures were used as input to train the ANN model and the performance is evaluated. The results show that the worm wheel is predominant for the fault over the other entities. ANN model predicts fault with an accuracy of 92.9%. Research outcomes envisage that the methodology used has good solidity to improve the performance.
Evaluation of performance of vibration signatures for condition monitoring of worm gearbox by using ANN
Int J Interact Des Manuf
Barshikar, Raghavendra R. (author) / Baviskar, Prasad R. (author)
2024-12-01
14 pages
Article (Journal)
Electronic Resource
English
Gearbox Condition Estimation Using Cyclo-Stationary Properties of Vibration Signal
| British Library Online Contents | 2009
Gearbox Condition Monitoring Based on Curvilinear Component Analysis
| British Library Online Contents | 2003