In English

Towards modelling of motion of ground vehicle with active suspension

Samu Almeida Chantre Fortes
Göteborg : Chalmers tekniska högskola, 2005. 51 s. Diploma work - Department of Applied Mechanics, Chalmers University of Technology, Göteborg, Sweden, ISSN 1652-8557; 2005:45, 2005.
[Examensarbete på avancerad nivå]

This master thesis work shows the capability of active and fully-active suspensions to improve the performance of a ground vehicle. The study is based on showing substantial vehicle improvement gains in comparison with the performance of standard vehicle suspension system. Ride comfort (sprung mass acceleration) and handling (tire displacement) are the two main criteria used to evaluate these improvements. Remarkable improvements over the passive suspension are demonstrated base on vehicle simulations and analyses. In this work, the quarter car model is used and has been implemented and simulated in two software environment, MSC.ADAMS and Matlab. The only input is the road profile that introduces vibration to the system. Two road profiles are used: 1) off road and 2) smooth road. First the passive suspension is simulated using the MSC.ADAMS model. At this stage of the work the response of the passive suspension according to road profiles is analyzed and improved in order to get the best suspension setting for each road profile. The next step is to simulate the passive suspension with the founded best settings using MatLab model. This procedure allows the validation of the results achieved with MSC.ADAMS model. Also the values achieved by others researchers are used to verify the accuracy of the results. At this stage of work important conclusion is that for each road profiles a specific suspension setting must be used if one wants to maintain good vehicle performance. Therefore “smart suspension” that can change their characteristic continuously is a priority in this work. Next step is the implementation of the active and fully-active suspension in MatLab model were control law is used to predict the actuator force. It is assumed that the system is linear; hence the control law is based on a linear state feedback controller where the Linear Quadratic Gaussian regulator, LQG, is used to calculate the feedback gain matrix minimizing the cost function used to calculate the actuator force. It is commonly assumed that the commanded force is produced accurately. Simulation is done without considering the actuators dynamics. The final step is to simulate and analyze the improvements brought by the active and fully-active suspensions. This part of work is performed using the Matlab. Again the best suspension settings found earlier for the two different roads are used in these simulations. For the fully-active suspension apart from the actuator only the spring coil is used whereas in the active suspension both the spring and damper are used in parallel with an actuator. As expected the results show considerable improvements of vehicle ride comfort particularly for the off road. The road handling improvement is minimal in both road profiles. In order to improve the handling these systems must suppress the motion created by the wheel-hop frequency situated at high frequencies values creating higher reaction forces deteriorating the improved ride comfort. Nevertheless, these systems still brings in improvements when particular driving condition arises, for example: reducing pitch motion in acceleration and braking, minimizing rollover in cornering and maintaining the ride height for payload compensation.



Publikationen registrerades 2006-01-19. Den ändrades senast 2013-04-04

CPL ID: 10478

Detta är en tjänst från Chalmers bibliotek