Prediction of Acoustic Natural Modes and Natural Frequencies Using Deep Learning

2021 ◽  
Vol 45 (12) ◽  
pp. 1137-1147
Author(s):  
Jae Ho Cho ◽  
Jin Woo Lee
Author(s):  
C. Rajalingham ◽  
R. B. Bhat ◽  
G. D. Xistris

Abstract The natural frequencies and natural modes of vibration of uniform elliptic plates with clamped, simply supported and free boundaries are investigated using Rayleigh-Ritz method. A modified polar coordinate system is used to investigate the problem. Energy expressions in Cartesian coordinate system are transformed into the modified polar coordinate system. Boundary characteristic orthogonal polynomials in the radial direction, and trigonometric functions in the angular direction are used to express the deflection of the plate. These deflection shapes are classified into four basic categories, depending on its symmetrical or antisymmetrical property about the major and minor axes of the ellipse. The first six natural modes in each of the above categories are presented in the form of contour plots.


Algorithms ◽  
2018 ◽  
Vol 11 (8) ◽  
pp. 112 ◽  
Author(s):  
Ruhua Wang ◽  
Ling Li ◽  
Jun Li

In this paper, damage detection/identification for a seven-storey steel structure is investigated via using the vibration signals and deep learning techniques. Vibration characteristics, such as natural frequencies and mode shapes are captured and utilized as input for a deep learning network while the output vector represents the structural damage associated with locations. The deep auto-encoder with sparsity constraint is used for effective feature extraction for different types of signals and another deep auto-encoder is used to learn the relationship of different signals for final regression. The existing SAF model in a recent research study for the same problem processed all signals in one serial auto-encoder model. That kind of models have the following difficulties: (1) the natural frequencies and mode shapes are in different magnitude scales and it is not logical to normalize them in the same scale in building the models with training samples; (2) some frequencies and mode shapes may not be related to each other and it is not fair to use them for dimension reduction together. To tackle the above-mentioned problems for the multi-scale dataset in SHM, a novel parallel auto-encoder framework (Para-AF) is proposed in this paper. It processes the frequency signals and mode shapes separately for feature selection via dimension reduction and then combine these features together in relationship learning for regression. Furthermore, we introduce sparsity constraint in model reduction stage for performance improvement. Two experiments are conducted on performance evaluation and our results show the significant advantages of the proposed model in comparison with the existing approaches.


2013 ◽  
Vol 12 (2) ◽  
pp. 205-212
Author(s):  
Daniel Burkacki ◽  
Michał Wójcik ◽  
Robert Jankowski

In technical branches, such as chemical or petroleum industries, cylindrical steel tanks are essential structures used for storage of liquid products. Therefore, their safety and reliability is essential, because any failure might have dangerous consequences, in extreme cases may even lead to an environmental disaster. The aim of the presented paper is to show the results of the modal analysis concerning the cylindrical steel tank with self-supported roof which has been constructed in northern Poland. The investigation was carried out with the use of the FEM commercial computer program Abaqus. The values of natural frequencies, as well as the natural modes, for different levels of liquid filling (empty tank, partly filled and tank fully filled) were determined in the study. The results of the study clearly indicate that the increase in the liquid level leads to the substantial decrease in the natural frequencies of the structure.


1952 ◽  
Vol 19 (4) ◽  
pp. 471-477
Author(s):  
Winston F. Z. Lee ◽  
Edward Saibel

Abstract A general expression is developed from which the frequency equation for the vibration of a constrained beam with any combination of intermediate elastic or rigid supports, concentrated masses, and sprung masses can be found readily. The method also is extended to the case where the constraint is a continuous elastic foundation or uniformly distributed load of any length. This method requires only the knowledge of the natural frequencies and natural modes of the beam supported at the ends in the same manner as the constrained beam but not subjected to any of the constraints between the ends. The frequency equation is obtained easily and can be solved to any desired degree of approximation for any number of modes of vibration in a quick and simple manner. Numerical examples are given for a beam with one concentrated mass, for a beam with one sprung mass, and a continuous beam with one sprung mass.


2013 ◽  
Vol 446-447 ◽  
pp. 597-601
Author(s):  
H. Haidzir ◽  
Dayang Laila Majid ◽  
A.S.M. Rafie ◽  
M.Y. Harmin

In any flutter prediction analysis, modal testing is necessary because flutter, a resonant like vibration occurs at a flutter frequency and adopts a mode shape akin to its structural natural modes. Modal testing can be performed computationally with knowledge of the mechanical properties of the structure. In the present work, computational modal analysis is first performed on a cantilevered hybrid composite thin plate and validated experimentally. Then, the computational procedure is demonstrated on a composite hollow wing model of same material. The concept of hollow wing is explored due to the superior mechanical properties of carbon/kevlar composite plate. It is observed that the natural frequencies of the hollow wing model are higher than thin plate due to stiffer configuration. A breathing mode was also observed at mode 4 for the hollow wing.


1992 ◽  
Vol 59 (4) ◽  
pp. 983-990 ◽  
Author(s):  
L. Weaver ◽  
L. Silverberg

This paper introduces node control, whereby discrete direct feedback control forces are placed at the nodes of the N+1th mode (the lowest N modes participate in the response). Node control is motivated by the node control theorem which states, under certain conditions, that node control preserves the natural frequencies and natural modes of vibration of the controlled system while achieving uniform damping. The node control theorem is verified for uniform beams with pinned-pinned, cantilevered, and free-free boundary conditions, and two cases of beams with springs on the boundaries. A general proof of the node control theorem remains elusive.


2008 ◽  
Vol 20 (1) ◽  
pp. 61-67
Author(s):  
Tohru Sasaki ◽  
◽  
Kunio Koizumi ◽  
Motofumi Sasaki

This paper deals with characteristics of the elliptic resonator. The resonator consists of loop and leg part, respectively. Each part was actuated by bimorph type piezoelectric member. The deflection of loop part and the bending of leg part generated by piezoelectric bimorphs makes an elliptic motion at the center point of the upper member. If resonators are put in a train and are actuated at the same frequency, it can feed objects on the upper surface of its train. There are two type of resonator, arch type and pentagonal type. Arch type resonator generates big amplitude on the center of upper surface. Bending its under member, pentagonal type resonator generates amplitude on the whole upper surface. Calculated results show that a shape of resonator influences on natural modes and natural frequencies. The trial one could generate elliptic motion in arbitrary phase with both type resonator. A pentagonal type resonator with notch reinforces its torsional stiffness and can generate necessary natural modes and an elliptic vibration without generating other natural modes.


Author(s):  
Costantino Carmignani ◽  
Paola Forte ◽  
Gabriele Melani ◽  
Giovanni Di Carlo

Aircraft transmissions have the peculiar characteristics of light structures and high operating speeds, therefore relatively low flexural natural frequencies and high excitation frequencies due to rotation and meshing. Thus some dynamic phenomena, such as traveling wave vibration (TWV), can be more easily encountered. However few papers can be found in the literature on TWV of gears while on the contrary it might be a problem especially for gears having thin rim and disk. This paper deals with the simulation of the disk TWV of bevel gears. 3D FEM is used to obtain the natural modes and frequencies of the gear. The response of a simplified non-rotating model of two meshing gears was entirely simulated in an Ansys environment. The numerical tests were performed considering a variable torque applied to one of the gears, the other being constrained, accounting for the meshing excitation. To simulate the rotating contact condition with respect to the gear reference systems, the dof of the nodes in contact of the two gears were coupled. Preliminary tests were performed considering one of the two gears rigid. The simulation of the TWV of gears makes it possible to identify the most critical conditions for stress and fatigue in the gear disks and therefore estimate the gear structural reliability.


2012 ◽  
Vol 516-517 ◽  
pp. 731-734 ◽  
Author(s):  
Wei Qiang Zhao ◽  
Yong Xian Liu ◽  
Mo Wu Lu

This paper introduces a FEA method for vibration characteristics analysis of blade-disc structure and makes a modal analysis of a certain gas turbine blade-disc based on this method. The natural frequencies and natural modes of this blade-disc are obtained and also vibration characteristics of each natural mode and corresponding influence on gas turbine blade-disc and other components are discussed in detail. The analysis method and analysis results presented in this paper are helpful for further research on optimal design and vibration safety verification for this gas turbine blade-disc.


Sign in / Sign up

Export Citation Format

Share Document