Failure prognostics of heavy vehicle hydro-pneumatic spring based on novel degradation feature and support vector regression

The hydro-pneumatic spring, as an important element of the suspension system for heavy vehicles, has attracted the attention of researchers for a long time because it plays such an important role in the steering stability, driving comfort, and driving safety of these vehicles. In this paper, we aim to solve the maintenance problems caused by gas leakage and oil leakage faults in hydro-pneumatic springs. The causes of hydro-pneumatic spring faults and their modes are investigated first. Then, we propose a novel time domain fault feature, called degraded pressure under the same displacement, and a novel feature extraction method based on linear interpolation and redefined time intervals. This feature extraction method is then combined with a data-driven prognostic method that is based on support vector regression to predict the failure trends. When compared with traditional prognostic methods for suspension systems based on vibration signals and vehicle dynamics models, the proposed method can evaluate the real-time spring condition without use of additional sensors or an accurate dynamic model. Therefore, the computational cost of the proposed method is very low and is also suitable for use in vehicles that are equipped with low-cost microprocessors. In addition, hydro-pneumatic spring performance degradation experiments and simulations based on AMEsim software are designed. The experimental data, real vehicle historical data, and simulation data are used to verify the feasibility of the proposed method.