Parameter Identification of a Driveline System by Deep Neural Networks
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20215334
- Paper/Info type
- Proceedings (Spring)
No.73-21
- Pages
- 1-7(Total 7 p)
- Date of publication
- May 2021
- Publisher
- JSAE
- Language
- English
- Event
- 2021 JSAE Annual Congress (Spring)[Online Meeting]
Detailed Information
Author(E) | 1) Davide Gorgoretti, 2) Wouter Vandermeulen, 3) Toshio Fuwa |
---|---|
Affiliation(E) | 1) Siemens Digital Industries Software, 2) Siemens Digital Industries Software, 3) Toyota Motor |
Abstract(E) | Identifying the physical parameters of a mechanical system is an essential process to correctly model its dynamics. Such identification usually relies on costly and time-consuming procedures, which may require both disassembly of system components and several physical measurements. This paper proposes a new methodology to carry out parameter identification in a fast and cost-effective way through Deep Neural Networks, which are trained to predict parameter values using data from the simulation model of the system under investigation. To highlight the viability of the proposed workflow, results are shown for a driveline system. |