Repository logo
 

Implications of nonlinear suspension behaviour for feedforward control of road noise in cars


Type

Thesis

Change log

Authors

Abstract

Active control offers a lightweight solution to the problem of low frequency structure-borne road noise in cars. Feedforward control of road noise is predominantly limited by the difficulty of finding reference signals that are highly coherent with the sound in the cabin. One factor that could restrict the coherence of the reference signals, and therefore the performance of a linear control system, is the behaviour of nonlinear components in the suspension. The purpose of this research is to identify potential sources of nonlinearity in a car’s suspension and investigate their influence on the propagation of structure-borne road noise into the cabin. The details of experiments and simulations are described that aim to quantify the role of nonlinearity in the suspension’s dynamics, and determine to what extent it limits the performance of a linear feedforward road noise control system. The results of experiments conducted on a Nissan Leaf test vehicle are presented that establish the theoretical maximum noise reduction that can be achieved with a linear control system using reference sensors placed at the wheel hubs. Laboratory experiments on the components in the Nissan Leaf’s suspension are presented that identify the hydraulic dampers as the main source of nonlinearity in the road noise transmission path. Models of the key components in a car suspension are developed based on the results of the experiments. Models for the front and rear dampers in the Nissan Leaf are validated up to 300 Hz, and analysed to reveal the mechanisms causing their nonlinear dynamics. Models for the bushings, springs and wishbone are developed and combined into a full suspension model. The suspension model’s predictions add to the evidence from the measurements that road noise transmission is strongly influenced by nonlinearity in the damper. Some implications of the suspension’s nonlinear dynamics for feedforward control of road noise are then explored. The linear control systems used currently are often limited to one reference sensor on each suspension; a simplified analysis of a two transmission path suspension system shows that, if one path is nonlinear, a linear control system with a single reference sensor can not cancel all of the noise generated at the wheel. This thesis presents strong evidence that nonlinear suspension dynamics limit the performance of feedforward road noise controllers. A nonlinear model of vibration transmission through a suspension is developed and guidelines are suggested for reference sensor selection.

Description

Date

2020-10-16

Advisors

Butlin, Tore
Langley, Robin

Keywords

road noise, nonlinear vibration, active control

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Sponsorship
Bose Corporation