• Communications for Statistical Applications and Methods
• /
• v.24 no.2
• /
• pp.143-154
• /
• 2017

A variable selection method based on probabilistic principal component analysis (PCA) using penalized likelihood method is proposed. The proposed method is a two-step variable reduction method. The first step is based on the probabilistic principal component idea to identify principle components. The penalty function is used to identify important variables in each component. We then build a model on the original data space instead of building on the rotated data space through latent variables (principal components) because the proposed method achieves the goal of dimension reduction through identifying important observed variables. Consequently, the proposed method is of more practical use. The proposed estimators perform as the oracle procedure and are root-n consistent with a proper choice of regularization parameters. The proposed method can be successfully applied to high-dimensional PCA problems with a relatively large portion of irrelevant variables included in the data set. It is straightforward to extend our likelihood method in handling problems with missing observations using EM algorithms. Further, it could be effectively applied in cases where some data vectors exhibit one or more missing values at random.

• Computers and Concrete
• /
• v.13 no.2
• /
• pp.187-207
• /
• 2014

This paper mainly discusses the influence of the aggregate properties including grading, shape, content and distribution on the chloride diffusion coefficient, as well as the initiation time of steel corrosion from a probabilistic point of view. Towards this goal, a simulation method of random aggregate structure (RAS) based on elliptical particles and a procedure of finite element analysis (FEA) at meso-scale are firstly developed to perform the analysis. Next, the chloride diffusion coefficient ratio between concrete and cement paste $D_{app}/D_{cp}$ is chosen as the index to represent the effect of aggregates on the chloride diffusion process. Identification of the random distribution of this index demonstrates that it can be viewed as actually having a normal distribution. After that, the effect of aggregates on $D_{app}/D_{cp}$ is comprehensively studied, showing that the appropriate properties of aggregates should be decided by both of the average and the deviation of $D_{app}/D_{cp}$. Finally, a case study is conducted to demonstrate the application of this mesoscopic method in predicting the initiation time of steel corrosion in reinforced concrete (RC) structures. The mesoscopic probabilistic method developed in this paper can not only provide more reliable evidences on the proper grading and shape of aggregates, but also play an important role in the probability-based design method.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.13 no.4
• /
• pp.39-48
• /
• 2005

Finite element analyses have been performed for the purpose of obtaining the robust and reliable design of engine cylinder block. Fatigue under high cycle operating loads is a primary concern and is evaluated by a probabilistic method. The robust and reliable design by a probabilistic method can provide satisfactory design conditions for the performance of the system under the influence of noise factors. Therefore, the design by this method will be desensitized to the uncontrollable noise factors. The simple methodology evaluates the distortion of main bore is proposed for the purpose of maintaining a well-controlled clearance between the crankshaft and main bores. The proposed methodology has proven a capability of predicting the distortion of the main bore under assembly, thermal, and firing loads. The calculated results are correlated well with the experimental ones.

• Nuclear Engineering and Technology
• /
• v.48 no.2
• /
• pp.545-553
• /
• 2016

The failure probabilities of the reactor pressure vessel (RPV) for low temperature over-pressurization (LTOP) and cool-down transients are calculated in this study. For the cool-down transient, a pressure-temperature limit curve is generated in accordance with Section XI, Appendix G of the American Society of Mechanical Engineers (ASME) code, from which safety margin factors are deliberately removed for the probabilistic fracture mechanics analysis. Then, sensitivity analyses are conducted to understand the effects of some input parameters. For the LTOP transient, the failure of the RPV mostly occurs during the period of the abrupt pressure rise. For the cool-down transient, the decrease of the fracture toughness with temperature and time plays a main role in RPV failure at the end of the cool-down process. As expected, the failure probability increases with increasing fluence, Cu and Ni contents, and initial reference temperature-nil ductility transition ($RT_{NDT}$). The effect of warm prestressing on the vessel failure probability for LTOP is not significant because most of the failures happen before the stress intensity factor reaches the peak value while its effect reduces the failure probability by more than one order of magnitude for the cool-down transient.

• Journal of KIISE
• /
• v.42 no.7
• /
• pp.868-877
• /
• 2015

The barcode area extraction from images has been extensively studied, and existing methods exploit frequency characteristics or depend on the Hough transform (HT). However, the slantedness of the images and noise affects the performance of these approaches. Moreover, it is difficult to deal with the case where an image contains multiple barcodes. We therefore propose a barcode detection algorithm that is robust under such unfavorable conditions. The pre-processing step implements a probabilistic Hough transform to determine the areas that contain barcodes with a high probability, regardless of the slantedness, noise, and the number of instances. Then, a frequency component analysis extracts the barcodes. We successfully implemented the proposed system and performed a series of barcode extraction tests.

• Structural Engineering and Mechanics
• /
• v.50 no.5
• /
• pp.697-708
• /
• 2014

The intentional buckling design of micro-films has various potential applications in engineering. The buckling amplitude and critical strain of micro-films are the crucial parameters for the buckling design. In the reported studies, the film parameters were regarded as deterministic. However, the geometrical and physical parameters uncertainty of micro-films due to manufacturing becomes prominent and needs to be considered. In the present paper, the probabilistic nonlinear buckling analysis of micro-films with uncertain parameters is proposed for design accuracy and reliability. The nonlinear differential equation and its asymptotic solution for the buckling micro-film with nominal parameters are firstly established. The mean values, standard deviations and variation coefficients of the buckling amplitude and critical strain are calculated by using the probability densities of uncertain parameters such as the film span length, thickness, elastic modulus and compressive force, to reveal the effects of the film parameter uncertainty on the buckling deformation. The results obtained illustrate the probabilistic relation between buckling deformation and uncertain parameters, and are useful for accurate and reliable buckling design in terms of probability.

• Geomechanics and Engineering
• /
• v.25 no.1
• /
• pp.17-30
• /
• 2021

This paper develops a convenient approach for deterministic and probabilistic evaluations of tunnel face stability using support vector machine classifiers. The proposed method is comprised of two major steps, i.e., construction of the training dataset and determination of instance-based classifiers. In step one, the orthogonal design is utilized to produce representative samples after the ranges and levels of the factors that influence tunnel face stability are specified. The training dataset is then labeled by two-dimensional strength reduction analyses embedded within OptumG2. For any unknown instance, the second step applies the training dataset for classification, which is achieved by an ad hoc Python program. The classification of unknown samples starts with selection of instance-based training samples using the k-nearest neighbors algorithm, followed by the construction of an instance-based SVM-KNN classifier. It eventually provides labels of the unknown instances, avoiding calculate its corresponding performance function. Probabilistic evaluations are performed by Monte Carlo simulation based on the SVM-KNN classifier. The ratio of the number of unstable samples to the total number of simulated samples is computed and is taken as the failure probability, which is validated and compared with the response surface method.

• Nuclear Engineering and Technology
• /
• v.53 no.3
• /
• pp.967-973
• /
• 2021

In 2011, an earthquake and subsequent tsunami hit the Fukushima Daiichi Nuclear Power Plant, causing simultaneous accidents in several reactors. This accident shows us that if there are several reactors on site, the seismic risk to multiple units is important to consider, in addition to that to single units in isolation. When a seismic event occurs, a seismic-failure correlation exists between the nuclear power plant's structures, systems, and components (SSCs) due to their seismic-response and seismic-capacity correlations. Therefore, it is necessary to evaluate the multi-unit seismic risk by considering the SSCs' seismic-failure-correlation effect. In this study, a methodology is proposed to obtain the seismic-response-correlation coefficient between SSCs to calculate the risk to multi-unit facilities. This coefficient is calculated from a probabilistic multi-unit seismic-response analysis. The seismic-response and seismic-failure-correlation coefficients of the emergency diesel generators installed within the units are successfully derived via the proposed method. In addition, the distribution of the seismic-response-correlation coefficient was observed as a function of the distance between SSCs of various dynamic characteristics. It is demonstrated that the proposed methodology can reasonably derive the seismic-response-correlation coefficient between SSCs, which is the input data for multi-unit seismic probabilistic safety assessment.

• Structural Engineering and Mechanics
• /
• v.81 no.4
• /
• pp.429-441
• /
• 2022

A probabilistic assessment of the seismic-excited buildings with a multiple-tuned-mass-damper (MTMD) system is carried out in the presence of uncertainties of the structural model, MTMD system, and the stochastic model of the seismic excitations. A free search optimization procedure of the individual mass, stiffness and, damping parameters of the MTMD system based on the snap-drift cuckoo search (SDCS) optimization algorithm is proposed for the optimal design of the MTMD system. Considering a 10-story structure in three cases equipped with single tuned mass damper (STMS), 5-TMD and 10-TMD, sensitivity analyses are carried out using Sobol' indices based on the Monte Carlo simulation (MCS) method. Considering different seismic performance levels, the reliability analyses are done using MCS and kriging-based MCS methods. The results show the maximum structural responses are more affected by changes in the PGA and the stiffness coefficients of the structural floors and TMDs. The results indicate the kriging-based MCS method can estimate the accurate amount of failure probability by spending less time than the MCS. The results also show the MTMD gives a significant reduction in the structural failure probability. The effect of the MTMD on the reduction of the failure probability is remarkable in the performance levels of life safety and collapse prevention. The maximum drift of floors may be reduced for the nominal structural system by increasing the TMDs, however, the complexity of the MTMD model and increasing its corresponding uncertainty sources can be caused a slight increase in the failure probability of the structure.

• The Korean Journal of Applied Statistics
• /
• v.36 no.4
• /
• pp.279-294
• /
• 2023

It is well known to generate synthetic data sets by the sequential regression multiple imputation (SRMI) method. The R-package synthpop are widely used for generating synthetic data by the SRMI approaches. In this paper, I suggest generating synthetic data based on the probabilistic principal component analysis (PPCA) method. Two simple data sets are used for a simulation study to compare the SRMI and PPCA approaches. Simulation results demonstrate that pairwise coefficients in synthetic data sets by PPCA can be closer to original ones than by SRMI. Furthermore, for the various data types that PPCA applications are well established, such as time series data, the PPCA approach can be extended to generate synthetic data sets.