Artificial Neural Network for Prediction of Seat-to-Head Frequency Response Function During Whole Body Vibrations in the Fore-and-Aft Direction
Vibrations while driving, regardless of their intensity and shape, have the most obvious effect of reducing driving comfort. Seat-to-head frequency response function (STHT) is a complex relationship resulting from the movement of the head due to the action of excitation on the seat in the form of vi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Faculty of Mechanical Engineering in Slavonski Brod, Faculty of Electrical Engineering in Osijek, Faculty of Civil Engineering in Osijek
2022-01-01
|
| Series: | Tehnički Vjesnik |
| Subjects: | |
| Online Access: | https://hrcak.srce.hr/file/412466 |
| Summary: | Vibrations while driving, regardless of their intensity and shape, have the most obvious effect of reducing driving comfort. Seat-to-head frequency response function (STHT) is a complex relationship resulting from the movement of the head due to the action of excitation on the seat in the form of vibrations in the seat/head interface. In this research, an artificial neural network model was developed, which aims to simulate the STHT function through the body of the subjects based on the data obtained experimentally. The experiments were conducted with twenty healthy male volunteers, who were exposed to single-axis fore-and-aft random broadband vibration. All the results of the experiment were recorded on the basis of which the artificial neural network (ANN) was trained. The developed ANN model has the ability to predict STHT values in the range of trained values both when changing the anthropometric measures of the subjects and changes in the input characteristics of vibrations. The mathematical models based on recurrent neural networks (RNN) used in this paper show with high accuracy STHT values in case there exists prior information about the anthropometric measures of the subjects and the input characteristics of vibrations. The results show that the expensive real-time simulations could be avoided by using reliable neural network models. |
|---|---|
| ISSN: | 1330-3651 1848-6339 |