• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.670-684
• /
• 2024

3D Cross-Modal Retrieval (3DCMR) is a task that retrieves 3D objects regardless of modalities, such as images, meshes, and point clouds. One of the most prominent methods used for 3DCMR is the Cross-Modal Center Loss Function (CLF) which applies the conventional center loss strategy for 3D cross-modal search and retrieval. Since CLF is based on center loss, the center features in CLF are also susceptible to subtle changes in hyperparameters and external inferences. For instance, performance degradation is observed when the batch size is too small. Furthermore, the Mean Squared Error (MSE) used in CLF is unable to adapt to changes in batch size and is vulnerable to data variations that occur during actual inference due to the use of simple Euclidean distance between multi-modal features. To address the problems that arise from small batch training, we propose a Noisy Center Loss (NCL) method to estimate the optimal center features. In addition, we apply the simple Siamese representation learning method (SimSiam) during optimal center feature estimation to compare projected features, making the proposed method robust to changes in batch size and variations in data. As a result, the proposed approach demonstrates improved performance in ModelNet40 dataset compared to the conventional methods.

• Architectural research
• /
• v.10 no.2
• /
• pp.21-26
• /
• 2008

The number of measured degrees of freedom for detecting the damage of any structures is usually less than the number of model degrees of freedom. It is necessary to expand the measured data to full set of model degrees of freedom for updating modal data. This study presents the expansion methods to estimate all static displacements and dynamic modal data of finite element model from the measured data. The static and dynamic methods are derived by minimizing the variation of the potential energy and the Gauss's function, respectively. The applications illustrate the validity of the proposed methods. It is observed that the numerical results obtained by the static approach correspond with the Guyan condensation method and the derived static and dynamic approaches provide the fundamental idea for damage detection.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.11a
• /
• pp.701-707
• /
• 2000

In this work, we propose to employ magnetostrictive sensors to develop a new non-contacting modal testing method. Specific applications are made in the modal testing of a beam in bending. The role of bias magnetic fields in measuring bending waves is addressed and an approximate analysis to explain the principle to measure bending signals is carried out. The measured modal data by the present method agree well with those by conventional methods using accelerometers.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.3 no.2
• /
• pp.147-154
• /
• 1999

A new approach to conducting the vibration parameters identification algorithm is proposed. The approach employs the concept of modal amplitude ratio implemented in a mode shape estimation. The accuracy of the improved Ibrahim Time Domain identification algorithm in extracting structural modal parameters from free response functions has been studied using computer simulated data for 9 stations on the two-span continuous beam. Simulated responses from the lumped and distributed parameter system demonstrate that this algorithm produces excellent results, even in the 300% noise response.

• Journal of the Society of Naval Architects of Korea
• /
• v.55 no.1
• /
• pp.37-44
• /
• 2018

To understand the dynamic characteristics of the vessel with hydroelastic response, it is very important to estimate the dynamic modal parameters such as mode shapes, natural frequency, and damping ratio. These dynamic modal parameters of full scale ship are a priori unknowns, hence to be estimated directly based upon the full scale measurement data. In this paper, dynamic modal parameters were extracted by signal processing of acceleration and strain data measured from a large container ship whose loading capacity is 9400TEU. The mode shapes of the vibrating hull were identified using the proper orthogonal decomposition and the vibration response of hull was decomposed into its modal magnitudes. Natural frequencies of specific modes were derived via Fourier transform of these modal magnitude. Also, the free decay signal of the vibrating hull was obtained through the random decrement technique and the damping ratio was estimated with accuracy.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.17 no.4
• /
• pp.389-403
• /
• 2004

This study presents the structural model modification based on the modal data such as natural frequencies and mode shapes. Preliminary structural model can be obtained using design drawings and field measurement, and therefore the deteriorated stiffness of a structure and the effect of the boundary conditions are difficult to be evaluated in preliminary analysis model, and the preliminary model can be modified using structural response data including static and/or dynamic characteristics. In this study, the structural model is modified based on the structural modal data using genetic algorithm. Modal testing were carried out for Imjin River Bridge and Hangjoo Bridge, the modal properties were estimated using modal identification techniques, and finally the structural models were updated using genetic algorithm. The modified structural model could give us more reliable structural analysis results and therefore those can be used for structural performance evaluation such as load carrying capacity and seismic capacity.

• Smart Structures and Systems
• /
• v.15 no.5
• /
• pp.1215-1232
• /
• 2015

In the present paper, a method for identifying damage in a multi storeyed shear building structure is presented using minimum number of modal parameters of the structure. A damage at any level of the structure may lead to a major failure if the damage is not attended at appropriate time. Hence an early detection of damage is essential. The proposed identification methodology requires experimentally determined sparse modal data of any particular mode as input to detect the location and extent of damage in the structure. Here, the first natural frequency and corresponding partial mode shape values are used as input to the model and results are compared by changing the sensor placement locations at different floors to conclude the best location of sensors for accurate damage identification. Initially experimental data are simulated numerically by solving eigen value problem of the damaged structure with inclusion of random noise on the vibration characteristics. Reliability of the procedure has been demonstrated through a few examples of multi storeyed shear structure with different damage scenarios and various noise levels. Validation of the methodology has also been done using dynamic data obtained through experiment conducted on a laboratory scale steel structure.

• Structural Monitoring and Maintenance
• /
• v.3 no.4
• /
• pp.395-406
• /
• 2016

Modal parameter identification has received much attention recently for their usefulness in earthquake engineering, damage detection and structural health monitoring. The identification method based on Matrix Pencil technique is adopted in this paper to identify structural modal parameters, such as natural frequencies, damping ratios and modal shapes using impulse vibration responses. This method can also be applied to dynamic responses induced by stationary and white-noise inputs since the auto- and cross-correlation function of the two outputs has the same form as the impulse response dynamic functions. Matrix Pencil method is very robust to noise contained in the measurement data. It has a lower variance of estimates of the parameters of interest than the Polynomial Method, and is also computationally more efficient. The numerical simulation results show that this technique can identify modal parameters accurately even if the noise level is high.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.32 no.5
• /
• pp.437-443
• /
• 2008

In this paper, analysis and control of the flexible multibody system using MATLAB is presented. The equations of motion of a flexible body are derived in terms of the modal coordinate. The rigid-flexible multibody dynamic solver is developed. Finite element information required to analyze motion of flexible bodies is imported from ANSYS. The modified finite element data, such as modal mass matrix, modal stiffness matrix and constraint mode shapes, is calculated in the solver. Since the solver is developed using MATLAB, it is very easy to connect with SIMULINK which is widely used to control motion of the multibody system. Several simulations are implemented to verify the developed solver. A control example is carried out and the usefulness of the developed solver is demonstrated.

• Structural Engineering and Mechanics
• /
• v.33 no.3
• /
• pp.325-337
• /
• 2009

Structural damage detection, damage localization and severity estimation of jacket platforms, based on calculating modal strain energy is presented in this paper. In the structure, damage often causes a loss of stiffness in some elements, so modal parameters; mode shapes and natural frequencies, in the damaged structure are different from the undamaged state. Geometrical location of damage is detected by computing modal strain energy change ratio (MSECR) for each structural element, which elements with higher MSECR are suspected to be damaged. For each suspected damaged element, by computing cross-modal strain energy (CMSE), damage severity as the stiffness reduction factor -that represented the ratios between the element stiffness changes to the undamaged element stiffness- is estimated. Numerical studies are demonstrated for a three dimensional, single bay, four stories frame of the existing jacket platform, based on the synthetic data that generated from finite element model. It is observed that this method can be used for damage detection of this kind of structures.