Determination of all acoustic material parameters of polymers II

点击翻译按钮获取中文摘要

Today, polymers are increasingly used in a wide field of applications. A typical example is the use of polymers in complex ultrasound-based measurement systems and devices. The development of such devices typically requires numerical simulations. Correct numerical predictions, however, can only be obtained if appropriate material models and parameters are available. Traditionally, the viscoelastic and temperature-dependent material properties of polymers are determined using quasi-static testing procedures, which are valid for low frequencies only. Ultrasound-based applications, however, require the knowledge of material parameters in the high-frequency range and thus the development of appropriate non-destructive methods for the determination of the latter with high precision.Previous collaborative work of the applicants has therefore been directed at the development of wave-based methods for material characterization. These are based on minimizing the discrepancy in measured and simulated signals using optimization techniques. Here, extruded cylindrical samples have been used. The corresponding material behavior of such samples can be idealized as transversally-isotropic. A significant increase in the sensitivity of the proposed algorithm with respect to the shear parameters has been obtained by using non-uniform excitations through a modified, segmented transducer setup. A specific numerical tool based on the scaled boundary finite element method (SBFEM) has been developed to minimize simulation time. The approach exploits not only the semi-analytical nature of the SBFEM but also the symmetric geometry of cylindrical samples. In addition, a significant gain in computational efficiency has been achieved by using differentiation of the proposed algorithm.The segmented transducer setup, however, is associated with additional challenges. In particular, precise alignment of transmitter and receiver is required to guarantee comparability with the numerical simulation. To reduce uncertainty, we aim to develop a new setup where only one transducer is attached to one face of the sample, acting as the transmitter and receiver. This modified setup will, in turn, result in higher requirements concerning signal processing and system characterization. Another aim of this project is to further improve the optimization algorithm for material characterization. Here, we plan to use the more robust cost function identified in the previous project to systematically optimize the hollow cylindrical geometry of samples such that unique material parameters are found. Measurements will be conducted and evaluated at varying temperatures in order to assess and validate the proposed methods. In this context, novel theoretical concepts of modeling damping will be compared to classical approaches and evaluated in terms of applicability.