31권 3호 191-197 2023 [한국자동차공학회 논문집 ]
| 제목 | PM10 비율 저감을 위한 인공신경망 활용 자동차용 브레이크 패드 마찰재 조성 최적화 |
|---|---|
| 분야 | 섀시 시스템 |
| 언어 | Korean |
| 저자 | 정성필(한국자동차연구원), 이필구(한국자동차연구원), 박재현(한국자동차연구원), 이영일(JBnl), 임수용(JBnl) |
| Key Words | Artificial neural network(인공신경망), Genetic algorithm(유전자 알고리즘), Regression model(회귀 모델), Bayesian optimization(베이지안 최적화), PM10 ratio(PM10 비율) |
| 초록 |
This study presents a methodology for optimizing the composition of brake pad materials by using an artificial neural network and a genetic algorithm. Through various regression models and an artificial neural network, a regression model that estimates the PM10 ratio change according to the composition of brake pad materials is derived, and the main parameters of each regression model are optimized through Bayesian optimization. By using a genetic algorithm, an optimized composition for each ingredient that minimizes the PM10 ratio is obtained, while maintaining the total ratio of ingredients at 100 %. An optimization verification is performed by using two types of test equipment: the chase friction test machine and the brake dynamometer show a 16.7 % and 10.7 % reduction of the PM10 ratio, respectively. Thus, the reliability of the optimization methodology presented in this study is verified. |
| 원문(PDF) | 다운로드 Journal Site |