Optimization of Quarter Car Suspension Dynamics Using Power Spectral Density of Irregular Road Profile

Vibrations are present in all types of vehicles and are very important for comfort and safety during travel. In this work, based on random road profiles obtained from the ISO 8608 standard of 2016, power spectral density (PSD) functions are obtained in terms of displacement and acceleration. Therewith, a methodology to obtain the variance and rms acceleration of the sprung mass is described. In sequence, it is proposed an ideal suspension design methodology that employs a multi-objective optimization technique based on Non-dominated Sorting Genetic Algorithm II (NSGA-II). In the computational implementation, the design criterion is defined as the minimization of the sprung mass vertical variance displacement, as well as the vertical rms acceleration of the sprung mass. Using a quarter car model, the damping and stiffness of the sprung mass are defined as design vectors. As a result, the NSGA II algorithm provides the Pareto front whose numerical values correspond to a set of feasible designs. Comparisons of the results with some methodologies described in the literature are made. The methodology proposed here leads to a decrease in the vibration amplitudes both in the frequency and time domains.