Adaptive Wavelet Neural Network Controller for Active Suppression Control of a Diaphragm-Type Pneumatic Vibration Isolator

Hung-Yi Chen,Jin-Wei Liang
DOI: https://doi.org/10.1007/s12555-014-0428-2
IF: 2.964
2017-01-01
International Journal of Control Automation and Systems
Abstract:This paper applies an intelligent control scheme to investigate a diaphragm-type pneumatic vibration isolation (PVI) problem. Adaptive wavelet neural network (AWNN) control is employed to control the PVI system which has inherited nonlinear and time-varying system characteristics. Since AWNN has excellent approximation capability originating from wavelet decomposition property and online learning ability stemming from neural networks, the method is especially suitable for the PVI control applications. In this paper, the adaptive learning rates are derived based on the Lyapunov stability theorem. Therefore, the stability of the closed-loop system can be assured. To validate the proposed method, a composite control scheme using pressure and velocity measurements as feedback signals is implemented. During experimental investigations, sinusoidal excitation with the excitation frequency close to the resonance and random-like signal are input on a floor base to simulate ground vibrations. Performances obtained from the proposed control scheme are compared with those obtained from passive isolation and PID scheme to illustrate the effectiveness of the proposed intelligent control.
What problem does this paper attempt to address?