Narrow-band and broadband vibration control of double-layer vibration isolation system based on a Youla parameterized adaptive controller

Abstract

Periodic motion equipment such as motors and pistons will generate broadband and narrow-band vibrations that vary with working conditions, thus reducing the process and measurement accuracy. The traditional control algorithm suppresses the vibration by increasing the broadband open-loop gain, but its effect on narrow-band vibration is limited. The Youla parameterized adaptive controller based on the internal model principle (IMP) is an effective method that can be used to suppress unknown time-varying narrow-band vibrations. However, the existing studies are insufficient for the control of broadband random vibrations. In this paper, a Youla parameterized adaptive controller with an improved central controller is designed to suppress broadband random and unknown narrow-band vibrations in a double-layer vibration isolation system. An optimal linear quadratic Gaussian (LQG) central controller is designed to first suppress the broadband random vibrations. Then, the global stable controllers with the Q parameter are obtained with the Youla parameterization method, and the Q parameter is adjusted online by the adaptive recursive least squares (RLS) algorithm to make the controller converge to the desired controller to suppress the unknown narrow-band vibrations. The results of experiments performed with a double-layer isolator show that the designed adaptive controller not only suppresses the broadband random vibrations to a certain extent but also almost completely eliminates the three unknown time-varying narrow-band vibrations. In practical applications, the controller can suppress the vibrations generated by a motor with varying rotational speeds, achieving an average peak attenuation of 95 dB with a tracking response time within 0.3 s.