The Robust H_∞ Control Study On Structure-Acoustic System

Active Structure Acoustic Control(ASAC) is the leading field in the area of Vibration and Noise Control. Based on the most current academic and experiment results of domestic and overseas in ASAC, modern robust control theory( H_∞control theory) and modern hydrokinetics theory,solid dynamics theory are used to build the control model of ASAC in this thesis. And the H_∞control experimental system is build based on modern signal processing technology and compter technology. The main contents in this thesis are as following:1. The feedback control model of structure-acoustic system is established and brought into'standard control framework'.2. The design problem of robust H_∞ASAC controller is researched for the first time in this thesis. The new method of design robust H_∞ASAC controller with mixed uncertainty is presented. The main ideal of the new method is as follows: the reduced modals are taken as additional disturbance for system, the original problem is transformed into the problem of robust performance design for the parameter uncertain system. Then the parameter uncertain problem is reduced to the standard H_∞control problem to solve. The mixed sensitivity problem is also researched in this thesis. The relation of the performance weighting function and the robust weighting function is analyzed. The effect of control path characteristic to system robust performance is taken as a new method to determine the locations of sensors and actuators.3. Under the conditions of sound velocity perturbation and the unmodel dynamic, active structure acoustic control system with robust H_∞controller is simulated in this thesis. The performance of the robust H_∞controller is compared with standard controller, the results show that the robust H_∞controller possess famous robust performance.4. A robust H_∞ASAC system for enclosed space structural sound radiation is build up for the first time, which is based on the TMS320C30 DSP board. The robust H_∞controller has been designed through the elections of the performance weighting function and the robust weighting function. The experiment results show that the single frequency noise can be reduced 7dB(A), and the broad-band random noise can be reduced effectively in assigned frequency band.