Modeling flexible plate structure system with free-free-clamped-clamped (FFCC) edges using particle swarm optimization

Abstract

This paper presents the performance of system identification for rectangular flexible plate using conventional parametric modeling approach by Recursive Least Squares (RLS) and intelligent parametric modeling approach by Particle swarm Optimization (PSO). The experimental rig of rectangular flexible plate with free-free-clamped-clamped edges boundary condition designed and fabricated in this research. The experimental study was conducted to collect the input-output data using a vibration flexible plate complete with data acquisition and instrumentation system. The input-output data will be used during developed the system identification. The whole developed model using RLS and PSO were validated using mean squares error (MSE), one step ahead prediction (OSA) and correlation test. The estimated of developed models was found are comparable, acceptable and possible to be used as a platform of controller development later on. As a comparison between developed system, it was found that the Particle Swarm Optimization algorithm has perform better in term of the lowest mean squares error which is 0.00032719.