Title: Neuro-genetic optimisation of disc brake speed sensitivity

Authors: Dragan Aleksendrić; Velimir Ćirović

Addresses: University of Belgrade Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade 11120, Serbia ' Innovation Center of Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade 11120, Serbia

Abstract: Since the driver obtains important feedback of a vehicle's dynamics and its braking capabilities depending on the change of brake performance, it represents an important aspect of a vehicle's performance and its quality of use. Sensitivity of braking torque vs. the friction couple interaction, under different braking conditions, is one of the most important properties of the disc brake. In this paper, we investigated possibilities for an intelligent dynamic optimisation of automotive braking system performance. The hybrid neuro-genetic optimisation model was developed in order to perform dynamic control and optimisation of the disc brake performance during a braking cycle. This model provided stabilisation of the brake performance and their maximisation vs. the brake pedal travel, selected by the driver, and the influence of the initial speed during a braking cycle. The model provided the disc brake actuation pressure which is adjusted upto the level that provides stable and, for current braking regimes, maximum braking torque values.

Keywords: vehicle design; vehicle braking; optimisation; disc brakes; brake performance; speed sensitivity; neural networks; genetic algorithms; vehicle dynamics; braking capabilities; braking torque; friction couple interaction; brake pedal travel; brake actuation pressure.

DOI: 10.1504/IJVD.2014.065716

International Journal of Vehicle Design, 2014 Vol.66 No.3, pp.258 - 271

Accepted: 26 Nov 2013
Published online: 03 Mar 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article