Vibration control of smart structures using optical fiber sensors and piezoelectric actuators

Abstract

Structural vibration suppression using optical fiber sensors and piezoelectric actuators has been investigated in order to develop the vibration suppression capability of fiber optic smart structures. The sensing capability of fiber optic smart structures has been demonstrated in a variety of engineering fields, accordingly fiber optic smart structures technology can provide us base structures with excellent sensing capability. Therefore, developing the vibration control function utilizing these integrated sensor networks is a practical way of developing multi-functional smart structures. Among several fiber optic sensor systems, extrinsic Fabry-Perot interferometer (EFPI) systems were selected as the main sensor systems, because the fabrication of an EFPI is rather easy and an EFPI sensor system can be constructed at relatively low price. First of all, the vibration control experiment has been performed without considering the sensor``s non-linearity. An optimal controller was designed and applied to the multi-mode vibration suppression. An EFPI sensor could be successfully applied to the vibration control if the vibration amplitude is small enough to produce approximately linear intensity signal. However, the output intensity of an EFPI sensor does not have linear relationship with mechanical strain because of its interferometric characteristics. The sensor``s non-linearity can affect a closed-loop system in its stability. So, the effects of interferometric non-linearity on the control performance and the stability of closed-loop system have been examined. A neural network based controller was applied to the vibration control of structural systems with sensor non-linearity. It is seen that a linear control system may make target system unstable when the sensor``s non-linearity gets high. With the minimum tuning of the control parameters, the neuro controller showed good performance and adaptiveness to the system with sensor non-linearity. However, there is a ...