Active structural-acoustic control of interior noise in a vibro-acoustic cavity incorporating system identification

Abstract

Linear quadratic Gaussian optimal control is one of the techniques used for active noise control. In practical implementation of this technique, one of the key difficulties faced is the estimation of the states of the plant. A state observer that accurately estimates these states can be used in this regard. Studies reported make use of analytically or experimentally derived models to build observers. This paper proposes a method for active noise control in the framework of active structural-acoustic control incorporating system identification for the development of the linear quadratic Gaussian controller. Kalman filter is used as a stochastic state observer of the plant states. System identification is carried out using modal testing and finite element model updating to obtain an accurate model of the plant for building up the Kalman filter. The objective of the proposed method is to actively reduce the noise inside the cavity due to disturbances acting on the cavity structure. The active control is achieved by controlling the structural vibrations by taking into account the degree of coupling between the various structural and the acoustic modes. The effectiveness of the proposed method is evaluated experimentally on a 3D rectangular box cavity with a flexible plate.