Artificial intelligence supported semi-active suspension system design for motorcycles


Creative Commons License

Beller S.

RESULTS IN ENGINEERING, cilt.28, ss.107551-107568, 2025 (ESCI, Scopus) identifier

  • Yayın Türü: Makale / Derleme
  • Cilt numarası: 28
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1016/j.rineng.2025.107551
  • Dergi Adı: RESULTS IN ENGINEERING
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
  • Sayfa Sayıları: ss.107551-107568
  • Anahtar Kelimeler: Motorcycle, Semi-active suspension system, Artificial neural networks, CFBNN, GRNN
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Çukurova Üniversitesi Adresli: Evet

Özet

Shock absorber adjustment on motorcycles is critical for ride comfort, road holding, and safety. Correctly

adjusted shock absorbers ensure that the motorcycle adapts optimally to the road conditions and make it easier

for the rider to maintain control. Shock absorbers absorb the energy produced by the springs and the impacts

they receive from the road and convert it into heat energy, which is transferred to the atmosphere and partly to

the chassis. In addition to the flow rate of the oil passing through holes of varying sizes, the thickness of the oil

also determines the movement speed of the shock absorber. The springs work in conjunction with the oil, and

their stiffness is sensitive to the load and the location of the shocks. Damping adjustment is made via holes or

valves and is sensitive to speed. A motorcycle with an incorrect shock absorber adjustment can be ridden, but if

the spring stiffness is not appropriate for the rider’s weight, riding will be impossible. In this study, a reservoir

shock absorber was subjected to a dynamic test with different spring stiffness settings and CST (Kinematic viscosity,

Centistokes mm2/s) values of silicone oils. During the experiments, different weights were dropped onto

the shock absorber from variable heights to simulate different dynamic driving conditions. Thus, the amounts

and durations of shock absorber sag and springback were examined. With the artificial intelligence-based control

block developed in the Simulink environment, different driving modes can be adjusted automatically or

manually according to variable road conditions.