Modelling and control of horizontal flexible plate using particle swarm optimization

Abstract

This paper presents the modeling and active vibration control using an evolutionary swarm algorithm known as particle swarm optimization. Initially, a flexible plate experimental rig was designed and fabricated with all clamped edges as boundary conditions constrained at horizontal position. The purpose of the experimental rig development is to collect the input-output vibration data. Next, the data acquisition and instrumentation system were designed and integrated with the experimental rig. Several procedures were conducted to acquire the input-output vibration data. The collected vibration data were then utilized to develop the system model. The parametric modeling using particle swarm optimization was devised using an auto regressive model with exogenous model structure. The developed model was validated using mean square error, one step ahead prediction, correlation tests and pole-zero diagram stability. Then, the developed model was used for the development of controller using an active vibration control technique. It was found that particle swarm optimization based on the active vibration control using Ziegler-Nichols method has successfully suppressed the unwanted vibration of the horizontal flexible plate system. The developed controller achieved the highest attenuation value at the first mode of vibration which is the dominant mode in the system with 34.37 dB attenuation.

Mayfly Algorithm for Modelling a Horizontal Flexible Plate Structure

“…In addition, the experimental plate rig was oriented horizontally to induce vertical vibrations onto the flexible plate. Complete specifications of the flexible plate utilized in this investigation are specified in Table 1 [6].…”

“…These accelerometers were directly linked to the data acquisition system (PCI 6259) located within the computer, connected to an SCC-68 via a shielded cable through the PCI-bus. A fully assembled experimental configuration utilized in this research is depicted in Figure 1 (b) [6].…”

“…The parametric modelling was carried out employing the RLS and MA, utilizing a linear auto-regressive with exogenous input (ARX) model structure. These models were formulated based on actual experimental vibration data acquired from the experimental setup [6].…”

“…In this study, the fitness function is defined by the minimum mean squared error (MSE) value as shown in Eq. ( 1) [6].…”

“…Therefore, to achieve this, the development of suitable models and effective control methods are essential, ensuring that flexible structures continue to play a significant role in various industries. Several approaches have been suggested by prior researchers and academic experts to address the concerns of undesired vibrations experienced by flexible plate structures [6]. Traditionally, passive vibration control (PVC) techniques have been employed to dampen excessive vibrations exerted towards such structures.…”