Application of AI-based method in vehicle NVH phenomena analysis
AIを活用した振動騒音現象の要因分析と応用
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20214940
- Paper/Info type
- Symposium Text
No.11-21
- Pages
- 1-6(Total 6 p)
- Date of publication
- Nov 2021
- Publisher
- JSAE
- Language
- Japanese
- Event
- JSAE Symposium 2021
Detailed Information
Category(J) | PPT資料 Translation |
---|---|
Category(E) | PPT slides |
Author(J) | 1) 山本 宇紘, 2) 小山 智樹, 3) 山下 寛子 |
Author(E) | 1) Takahiro Yamamoto, 2) Tomoki Koyama, 3) Hiroko Yamashita |
Affiliation(J) | 1) 株式会社SUBARU, 2) 株式会社SUBARU, 3) 株式会社SUBARU |
Affiliation(E) | 1) SUBARU Corp., 2) SUBARU Corp., 3) SUBARU Corp. |
Abstract(J) | 近年駆動系の構造や制御の複雑化が進む中、従来の手法でノイズ対策を行おうとすると、起振源や伝達経路の解明に必要な計測データの膨大化が著しい。そこで、統計的因果探索手法(LiNGAM)を用いて計測データの因果関係を推定し、分析の効率化を実現するとともに、CVTからの様々な発生音を対象に分析を行った。 Translation |
Abstract(E) | The development of devices and controls in transmission causes the complex and difficulty of the mechanism analysis of NVH phenomena. Clarifying the source of noise and the transfer paths requires a huge number of measured data and enormous man-hours of skilled engineers. This article shows the efficient method to analyze the causal structure of measured data; Linear Non-Gaussian Acyclic Model (LiNGAM). The article provides the practical application example of the CVT NVH analysis also. |