CONDITIONS FOR COMPLETE FEEDBACK LINEARISATION ON ACTIVE SUSPENSION SYSTEMS FOR MINING TRUCKS

摘要

This paper presents the design and implementation of a mining truck half-car active vehicle suspension system with a pre-disturbance detection algorithm that aims at reducing driver experienced vibrations. It addresses the complexities and limitations of input output feedback linearisation by addressing challenges of hidden state analysis involving jacobian invertibility and zero dynamics analysis on suspension models. The novelty of this paper entails adopting input state linearisation conditions to develop a set of states that result in complete linearisation having no hidden states. Negative state feedback controller is applied to the linearised system and the results demonstrate improved attenuated displacement and smoother accelerations hence cushioning the sprung load. A test bed is developed and integrated with the controller through the use of Matlab, Simulink and DSpace to demonstrate the active suspension system prototype results. Respectively 81.82 % and 97.5 % displacement and acceleration attenuation are achieved by the implemented linear controller.