Optimization of electrically interconnected suspension for vibration control
Publication Name
IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
Abstract
This paper establishes a 7-degree-of-freedom full-vehicle suspension system with two 2-degree-of-freedom controllable electrically interconnected suspensions (EIS) to achieve the control effect of the entire vehicle. The motions between the sprung mass and unsprung masses are converted into electrical energy via the electromagnetic suspensions, and then the energy is electrically interconnected with the electrical network (EN). This research is based on the previous study of a 2-degree-of-freedom EIS system, which shows the vehicle performance at heave and roll is improved by controlling resistances in the EN. On this basis, this project utilizes the genetic algorithm (GA) and regards the parameters of variable resistors in EN as the optimizing targets. Based on the factors of vehicle comfort and suspension deflection limits, the 11 control objectives have been defined in order to achieve optimal performance through GA optimization. The final parameters of the resistors R_e and R_l in the EN are obtained as 14.78\Omega and 7.29\Omega, respectively. Both the time-domain simulation and the frequency weighted root mean square analyses show that the three sprung mass accelerations of the full vehicle have been significantly improved. The optimized vertical heave acceleration has decreased by 48.13%, while the roll and pitch angular acceleration has also decreased by 34.91% and 49.76%, respectively. These outcomes have sufficiently proved the effectiveness of GA optimization for the vehicle model. Furthermore, the EIS system can also benefit from the reduction of motion sickness; the optimized motion sickness dosing value has reduced almost 45.43% compared with the initial value.
Open Access Status
This publication is not available as open access
Volume
2021-September
First Page
3890
Last Page
3895
Funding Number
DP200100149
Funding Sponsor
Australian Research Council