학술논문
Optimized Placement Criteria of Hybrid Mounting System on Smart Composite Structures for Active Vibration Control
Document Type
Dissertation/ Thesis
Source
Subject
Language
English
Abstract
Electric and hybrid vehicle engines produce a complex spectrum of vibration and noise. This tendency is observed especially for mid-frequency components. Various hybrid mounting techniques have been developed to isolate them. These are designed to continuously control the dynamic characteristics of the mounts and improve the noise, vibration, and harshness (NVH) performance under various operating conditions. Hybrid mounts have attracted attention as replacement for existing mounts by simultaneously realizing static and dynamic stiffness, which is important for supporting an engine. When hybrid engine mounting system placed in optimized location, the performance of the overall vibration reduction is more effective. Thus, in this study, in order to find the optimized placement criteria of hybrid engine mounting system on smart composite structures, the simulation of static and dynamic has been performed. Furthermore, the feasibility experiment is carried out to validate the proposed criteria.Overall smart composite structures having two hybrid mounting system has been modeled based on lumped parameter model. The control forces on each hybrid mounting system are calculated though static and dynamic method. According to change the position of two hybrid mounting system, the optimized placement criteria was determined when the minimum control force on the two hybrid mounting system. In addition, in order to confirm the simulation results, it has been carried out the feasibility experiment to validate that the vibration is greatly reduced at the optimal location.And sinusoids, amplitude modulated and frequency modulated signals as input signal applied to the previously verified positions on the smart composite structures, the adaptive filter is applied for control, and the normalization least mean square (NLMS) algorithm, which is commonly used in research, is extended to a Multi-NLMS algorithm. It is shown that when multi-frequency inputs are applied, higher attenuation possible at the targeted frequency.