Investigation Of Active Vibration Control With Online Secondary Path Modeling Based On Improved Gradient Algorithm

This paper proposes a new method for online secondary path modeling in active vibration control systems. To cancel interactions between the active control process and the modeling process greatly, the existing methods for active vibration control with online secondary path modeling consist of three adaptive filters. The main feature of the proposed method is that it uses only two adaptive filters. In the proposed method, a new type of improved gradient LMS algorithm is used. Most existing LMS algorithms are based on noisy gradient estimation since instantaneous squared errors are used to estimate ensemble average of squared errors. The gradient estimation in the proposed algorithm is improved by using the weighted time average instead of the instantaneous value to approximate the ensemble average. It can eliminate the interaction between the processes of online secondary path modeling and active control.A new adaptive strategy with online secondary path modeling based on improved gradient LMS algorithm is adopted in an active vibration control system. Comparisons are made between the present method and the existing ones by simulation in Matlab and experimental investigation for an isolation rig. The results of simulation and experiments demonstrate that good control performances are achieved in a broadened frequency range and the control system is robust.