• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2012.10a
• /
• pp.671-672
• /
• 2012

• Journal of the Korea Society for Simulation
• /
• v.32 no.1
• /
• pp.23-34
• /
• 2023

The purpose of this paper is to review global stochastic optimization algorithms(GSOA) in case binary response experimentation is used and to compare the performances of them. GSOAs utilise estimator of probability of success $\^p$ instead of population probability of success p, since p is unknown and only known by its estimator which has stochastic characteristics. Hill climbing algorithm algorithm, simple random search, random search with random restart, random optimization, simulated annealing and particle swarm algorithm as a population based algorithm are considered as global stochastic optimization algorithms. For the purpose of comparing the algorithms, two types of test functions(one is simple uni-modal the other is complex multi-modal) are proposed and Monte Carlo simulation study is done to measure the performances of the algorithms. All algorithms show similar performances for simple test function. Less greedy algorithms such as Random optimization with Random Restart and Simulated Annealing, Particle Swarm Optimization(PSO) based on population show much better performances for complex multi-modal function.

• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.905-910
• /
• 2004

Vibration characteristics of the FIV test loop for the Flow-Induced Vibration(FIV) study of a PWR partial(5x5) fuel assembly are investigated by the Finite Element(FE) analysis and the modal test. For the FE analysis, 3-D beam element is used for the pipes and the test section and mass element used for the valves and flanges. The 'U' restrainer stiffness determined by numerical simulation is used for the FE model. The result of the FE analysis is compared with that of the modal test. The higher mode similarity between the test and analysis is observed in a few low modes. After that, the mode similarity reduce as the mode goes high. It is concluded that the first to the third vibration modes are observed at the lower parts of the 6 inches restoring line, followed by a local mode at the test section, and the natural frequencies of the modes are 22.4 Hz, 26.0 Hz, 27.5 Hz and 31.4 Hz.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.665-668
• /
• 2008

The load distribution to each mode of a structure under seismic loading depends on the modal participation factor. The factor of an idealized analytical model, however, is different to the actual one due to modeling and construction error. Therefore, there exist limits on the estimation of actual behavior. In this study, an identification procedure for participation factor based on vibration test is proposed. The procedure has an advantage that the mode shape vector can also be estimated directly from the participation factor. The numerical simulation using a three story building is performed to evaluate the proposed procedure.

• Nuclear Engineering and Technology
• /
• v.55 no.3
• /
• pp.947-957
• /
• 2023

Curved cylindrical structures such as elbows have a non-uniform thickness distribution due to their fabrication process, and as a result have a number of complex mode shapes, including circumferential and axial nodal patterns. In nuclear power plants, material degradation is induced in pipes by flow accelerated erosion and corrosion, causing the wall thickness of carbon steel elbows to gradually thin. The corresponding frequencies of each mode shape vary according to the wall thinning state. Therefore, the thinning state can be estimated by monitoring the varying modal characteristics of the elbow. This study investigated the varying modal characteristics of artificially thinned carbon steel elbows for each thinning state using numerical simulation and experimental methods (MRIT, Multiple Reference Impact Test). The natural frequencies of specified mode shapes were extracted, and results confirmed they linearly decreased with increasing thinning. In addition, by comparing single FRF (Frequency Response Function) data with the results of MRIT, a concise and cost effective thinning estimation method was suggested.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2005.05a
• /
• pp.124-127
• /
• 2005

As computer capability and test skill become more and more advanced, finite element method and modal test are being widely applied in engineering design. In order to correlate and reconcile the inevitable discrepancies between the analytical and experimental models, many techniques have been developed. Among these methods, multiple-system methods are known as the effective tools in that they can supply the rich modal data available which are experimentally obtained. These abundant modal data can help structural system parameters estimated well. Multiple-system methods can be classified into the structural modification methods and feedback controller methods. The structural modification methods need the physical attachment of structures and their concept may limit the application of them. To overcome this drawback, the feedback controller methods are addressed which enable us to get more modal data without the structural change. Mode decoupling controller(MDC), one of them, is to use acceleration out)ut feedback to perturb an open-loop system. The output feedback controller generally cannot guarantee the stability of a closed-loop system. However, MDC can solve this problem under the certain constraints. So far, MDC utilizes accelerations as the sensor signals. In this research, strain sensors are going to be picked up to apply to the MDC. Strain output is recently used for structural system identification due to the drastically improved and miniaturized strain sensor. In this paper, we show that the MDC using strain output has differences compared with acceleration output in estimating the structural system parameters. The associated simulation is performed to demonstrate the above mentioned characteristics.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.16 no.3 s.108
• /
• pp.233-239
• /
• 2006

The effect of the joint condition of automotive muffler shell and baffle on radiated noise from muffler shell Is studied. Shell vibration characteristics, especially shift In natural frequency and change in of response amplitude, was identified from finite element analysis and modal tests for various joint conditions between shell and baffle. When there is heat expansion and inner pressure from the exhaust 9as in the muffler with elliptical cross section, the shell expands out, and construct gap with baffles. This gap causes more degrees of freedom of shell vibration and natural frequency shift. By welding the muffler shell to the baffles or using a dual shell muffler, the radiated noise problem could be mitigated supplying more a rigid joint and damping, respectively.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.25 no.12
• /
• pp.866-873
• /
• 2015

The transverse vibration of the deploying beam from rigid hub was analyzed by simulation and experiment. The linear governing equation of the deploying beam was obtained using the Euler-Bernoulli beam theory. To discretize the governing equation, the Galerkin method was used. After transforming the governing equation into the weak form, the weak form was discretized. The discretized equation was expressed by the matrix-vector form, and then the Newmark method was applied to simulate. To consider the damping effect of the beam, we conducted the modal test with various beam length. The mass proportional damping was selected by the relation of the first and second damping ratio. The proportional damping coefficient was calculated using the acquired natural frequency and damping ratio through the modal test. The experiment was set up to measure the transverse vibration of the deploying beam. The fixed beam at the carriage of the linear actuator was moved by moving the carriage. The transverse vibration of the deploying beam was observed by the Eulerian description near the hub. The deploying or retraction motion of the beam had the constant velocity and the velocity profile with acceleration and deceleration. We compared the transverse vibration results by the simulation and experiment. The observed response by the Eulerian description were analyzed.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1998.04a
• /
• pp.244-249
• /
• 1998

The present paper improves the direct identification scheme based upon the equation error formulation with incomplete modal data. First, an indirect estimation technique is considered for estimating unmeasured elements of latent vectors by the combined use of a model and measured incomplete eigen vectors. It is used for estimating the other elements of eigen vectors, which are essential for identification but not available. Next an index is introduced here to indicate the quality of estimation with respect to the mode and the measured positions. A sensitivity formula for eigenvalues with respect to the unknown joint coefficient is also derived to select the modes appropriate for identification. An identification strategy is suggested to meet with practical problems with the help of the index and sensitivity formula. The index and the sensitivity are proved to be useful for selecting measurement positions and modes appropriate for identification A comprehensive simulation study is performed to test the proposed method.

• Nuclear Engineering and Technology
• /
• v.55 no.3
• /
• pp.827-838
• /
• 2023

Centrifugal pump is a key part of nuclear power plant systems, and its health status is critical to the safety and reliability of nuclear power plants. Therefore, fault diagnosis is required for centrifugal pump. Traditional fault diagnosis methods have difficulty extracting fault features from nonlinear and non-stationary signals, resulting in low diagnostic accuracy. In this paper, a new fault diagnosis method is proposed based on the improved particle swarm optimization (IPSO) algorithm-based variational modal decomposition (VMD) and relevance vector machine (RVM). Firstly, a simulation test bench for rotor faults is built, in which vibration displacement signals of the rotor are also collected by eddy current sensors. Then, the improved particle swarm algorithm is used to optimize the VMD to achieve adaptive decomposition of vibration displacement signals. Meanwhile, a screening criterion based on the minimum Kullback-Leibler (K-L) divergence value is established to extract the primary intrinsic modal function (IMF) component. Eventually, the factors are obtained from the primary IMF component to form a fault feature vector, and fault patterns are recognized using the RVM model. The results show that the extraction of the fault information and fault diagnosis classification have been improved, and the average accuracy could reach 97.87%.