• Journal of the Earthquake Engineering Society of Korea
• /
• v.28 no.2
• /
• pp.113-119
• /
• 2024

Existing reinforced concrete buildings with seismically deficient column details affect the overall behavior depending on the failure type of column. This study aims to develop and validate a machine learning-based prediction model for the column failure modes (shear, flexure-shear, and flexure failure modes). For this purpose, artificial neural network (ANN), K-nearest neighbor (KNN), decision tree (DT), and random forest (RF) models were used, considering previously collected experimental data. Using four machine learning methodologies, we developed a classification learning model that can predict the column failure modes in terms of the input variables using concrete compressive strength, steel yield strength, axial load ratio, height-to-dept aspect ratio, longitudinal reinforcement ratio, and transverse reinforcement ratio. The performance of each machine learning model was compared and verified by calculating accuracy, precision, recall, F1-Score, and ROC. Based on the performance measurements of the classification model, the RF model represents the highest average value of the classification model performance measurements among the considered learning methods, and it can conservatively predict the shear failure mode. Thus, the RF model can rapidly predict the column failure modes with simple column details.

• Journal of the Korean Society for Railway
• /
• v.8 no.2
• /
• pp.170-175
• /
• 2005

This study presents a prediction of a failure rate in a safety required system that consists of a embedded control system, requiring a satisfaction of a quantitative safety requirement. International Standards are employed to achieve a regular procedures in the whole life cycle of a system, for the purpose of a prediction and a evaluation of a fault that might be able to be happened in a system. This International Standards uses SIL (Safety Integrity Level) to evaluate a safety level of a system. SIL is divided into 4 levels, from level 1 to level 4, and each level has functional failure rate and dangerous failure rate of a system. In this paper we describe the conventional method to predict the dangerous failure rate and propose a method using hazard analysis to predict the dangerous failure rate. The conventional method and the technique using hazard analysis to predict the dangerous failure rate are made a comparison through the control modules of the interlocking system in KTX. The proposed method verify better effectiveness for the prediction of the dangerous failure rate than that of the conventional method.

• Structural Engineering and Mechanics
• /
• v.66 no.4
• /
• pp.543-555
• /
• 2018

Fiber Reinforced Polymer (FRP), which has a high strength to weight ratio, are now regularly used for strengthening of deficient reinforced concrete (RC) structures. While various researches have been conducted on FRP strengthening, an area that still requires attention is predicting the debonding failure load of prestressed FRP strengthened RC beams. Application of prestressing increases the capacity and reduces the premature failure of the beams largely, though not entirely. Few analytical methods are available to predict the failure loads under flexure failure. With this paucity, this research proposes a method for predicting debonding failure induced by intermediate crack (IC) for prestressed FRP-strengthened beams. The method consists of a numerical study on beams retrofitted with prestressed FRP in the tension side of the beam. The method applies modified Branson moment-curvature analysis together with the global energy balance approach in combination with fracture mechanics criteria to predict failure load for complicated IC-induced failure. The numerically simulated results were compared with published experimental data and the average of theoretical to experimental debonding failure load is found to be 0.93 with a standard deviation of 0.09.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2002.05a
• /
• pp.89-92
• /
• 2002

Traditionally unidirectional laminated composite which are characterized by high specific stiffness and strength were used for structural application. But theses composites are highly susceptible to impact damage because of lower transverse tensile strength. The main failure modes of laminated composite are fiber breakage, matrix cracking and delamination for low velocity impact. The modified failure criterions are implemented to predict these failure modes with finite element analysis. Failure behavior of the woven fabric laminated composite which is used in forehead part of subway to lighten weigh has been studied. The new failure criterions are in good agreement with experimental results and can predict the failure behavior of the woven fabric composite plate subjected to low velocity impact more accurately.

• Proceedings of the Korean Geotechical Society Conference
• /
• 1994.09a
• /
• pp.231-234
• /
• 1994

Concentrated stresses due to the tunnel excavation easily cause failure around opening in the soft rock mass layer. Thus, while excavatng tunnel in the soft rock mass layerm it is very important to predict the possibility of failure or yielding zones around tunnel boundary. There are two typical methods to predict these; 1) the analysis of field monioring data and 2) numerical analysis. In this study, it was attempted to describe the time-dependent or progressive rock mass manner due to the continuous failure and fracturing caused by surrounding underground openings using the second method. In order to apply the effects of progressive failure underground, an iterative technique was used with the Hoek and Brown rock mass failure theory. By developing and simulating, three different shapes of twin tunnels, this research simulated and estimated the proper size of critical pillar width between tunnels, distributed stresses on the tunnel sides, and convergences of tunnel crowns. Moreover, results out progressive failure technique based on the Hoek and Brown theory were compared with the results out of Mohr-Coulomb theory.

• Nuclear Engineering and Technology
• /
• v.41 no.5
• /
• pp.715-722
• /
• 2009

A model to predict failure of designed-to-fail (dtf) fuel particles is discussed. The dtf fuel under study consisted of a uranium oxycarbide kernel coated with a single pyrocarbon seal coat. Coating failure was assumed to be due to fission gas recoil and knockout mechanisms and direct diffusive release of fission gas from the kernel, which acted to increase pressure and stress in the pyrocarbon layer until it ruptured. Predictions of dtf fuel failure using General Atomics' particle fuel performance code for HRB-17/18 and HFR-B1 irradiation tests were reasonably accurate; however, the model could not predict the failure for COMEDIE BD-1. This was most likely due to insufficient information on reported particle fuel failure at the beginning.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2002.10a
• /
• pp.513-520
• /
• 2002

The stability analysis of rock slope can be predicted using a suitable field data but it cannot be predicted unless suitable field data was taken. In this study, artificial neural networks theory is applied to predict plane failure that has a few data. It is well known that human brain has the advantage of handling disperse and parallel distributed data efficiently. On the basis of this fact, artificial neural networks theory was developed and has been applied to various fields of science successfully In this study, error back-propagation algorithm that is one of the teaching techniques of artificial neural networks is applied to predict plane failure. In order to verify the applicability of this model, a total of 30 field data results are used. These data are used for training the artificial neural network model and compared between the predicted and the measured. The simulation results show the potentiality of utilizing the neural networks for effective safety factor prediction of plane failure. In conclusion, the well-trained artificial neural network model could be applied to predict the plane failure stability of rock slope.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.5
• /
• pp.675-682
• /
• 2013

Reliability analysis and prediction of next failure time is critical to sustain weapon systems, concerning scheduled maintenance, spare parts replacement and maintenance interventions, etc. Since 1981, many methodology derived from various probabilistic and statistical theories has been suggested to do that activity. Nowadays, many A.I. tools have been used to support these predictions. Support Vector Regression(SVR) is a nonlinear regression technique extended from support vector machine. SVR can fit data flexibly and it has a wide variety of applications. This paper utilizes SVM and SVR with combining time series to predict the next failure time based on historical failure data. A numerical case using failure data from the military equipment is presented to demonstrate the performance of the proposed approach. Finally, the proposed approach is proved meaningful to predict next failure point and to estimate instantaneous failure rate and MTBF.

• Journal of the Korean Society of Safety
• /
• v.18 no.1
• /
• pp.116-122
• /
• 2003

Monte Carlo method is one of the powerful engineering tools especially to solve the complex non-linear problems. The Monte Carlo method gives approximate solution to a variety of mathematical problems by performing statistical sampling experiments on a computer. One of the methods to predict the time dependent failure probability of one of the bridge components or the bridge system is a lifetime function. In this paper, FORTRAN program is developed to predict the failure probability of bridge components or bridge system by using both system reliability and lifetime function. Monte Carlo method is used to generate the parameters of the lifetime function. As a case study, the program is applied to the concrete-steel bridge to predict the failure probability.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.4
• /
• pp.755-764
• /
• 2011

In many companies field failure data is used to predict the future number of failures, especially when an unexpected failure mode happens to be a problem. It is because they want to predict the number of spare parts needed and the future quality warranty cost associated with the part based on the predictions of the future number of failures. In this paper field summary data is used to predict the future number of failures based on an appropriate distribution. Other types of data are also investigated to identify the appropriate distribution.