Ardi Soma, Dirman Hanafi, Hisyam Abdul Rahman, Azriyenni Azhari Zakri, Riki Mukhaiyar, Setiyo Budiyanto
System identification plays a crucial role in modeling vehicle suspension dynamics to improve ride comfort and safety. This paper presents the application of Artificial Neural Networks (ANN) in identifying the nonlinear dynamics of a quarter-car passive suspension system. A Nonlinear AutoRegressive with eXogenous Inputs (NARX) model was employed to capture system behavior based on experimental input-output data collected from a test vehicle using LVDT and accelerometer sensors. Several activation function configurations were evaluated with Levenberg-Marquardt optimization for weight updates. The results show that the ANN-based approach effectively identifies the nonlinear dynamics of the quarter-car suspension model. At $5 \text{km} / \mathrm{h}$, the PurelinTansig activation achieved the lowest mean squared error (MSE) of 0.6023, while at higher speeds $(30-60 \text{km} / \mathrm{h})$, Logsig-Purelin and Radbas-Purelin maintained MSE values below 0.7. Overall, the ANN model achieved up to 21% lower error compared to other tested activation function combinations. The proposed method demonstrates significant potential for accurate suspension modeling, providing a foundation for future control system design and optimization. © 2025 IEEE.
PT. NH Kaffa Indonesia, Batam, Indonesia; Universiti Tun Hussein Onn Malaysia, Instrumentation and Sensing Technology (InSeT), Faculty of Electrical and Electronic Engineering, Batu Pahat, Malaysia; Universitas Pembangunan Nasional Veteran Jakarta, Faculty of Engineering, Dept. Electrical Engineering, Jakarta, Indonesia; Universiti Tun Hussein Onn Malaysia, Faculty of Electrical and Electronic Engineering, Dept. Electronic Engineering, Batu Pahat, Malaysia; Universitas Riau, Faculty of Engineering, Dept. of Electrical Engineering, Pekanbaru, Indonesia; Universitas Negeri, Faculty of Engineering, Dept. Electrical Engineering, Padang, Indonesia; Universitas Mercu Buana, Dept. of Electrical Engineering, Jakarta, Indonesia
