• Journal of Korean Association for Spatial Structures
• /
• v.23 no.3
• /
• pp.87-94
• /
• 2023

Chloride is one of the most common threats to reinforced concrete (RC) durability. Alkaline environment of concrete makes a passive layer on the surface of reinforcement bars that prevents the bar from corrosion. However, when the chloride concentration amount at the reinforcement bar reaches a certain level, deterioration of the passive protection layer occurs, causing corrosion and ultimately reducing the structure's safety and durability. Therefore, understanding the chloride diffusion and its prediction are important to evaluate the safety and durability of RC structure. In this study, the chloride diffusion coefficient is predicted by machine learning techniques. Various machine learning techniques such as multiple linear regression, decision tree, random forest, support vector machine, artificial neural networks, extreme gradient boosting annd k-nearest neighbor were used and accuracy of there models were compared. In order to evaluate the accuracy, root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R2) were used as prediction performance indices. The k-fold cross-validation procedure was used to estimate the performance of machine learning models when making predictions on data not used during training. Grid search was applied to hyperparameter optimization. It has been shown from numerical simulation that ensemble learning methods such as random forest and extreme gradient boosting successfully predicted the chloride diffusion coefficient and artificial neural networks also provided accurate result.

• Smart Structures and Systems
• /
• v.26 no.3
• /
• pp.345-360
• /
• 2020

Modern swarm intelligence heuristic search methods are widely applied in the field of structural health monitoring due to their advantages of excellent global search capacity, loose requirement of initial guess and ease of computational implementation etc. To this end, a hybrid strategy is proposed based on butterfly optimization algorithm (BOA) and differential evolution (DE) with purpose of effective combination of their merits. In the proposed identification strategy, two improvements including mutation and crossover operations of DE, and dynamic adaptive operators are introduced into original BOA to reduce the risk to be trapped in local optimum and increase global search capability. The performance of the proposed algorithm, hybrid butterfly optimization and differential evolution algorithm (HBODEA) is evaluated by two numerical examples of a simply supported beam and a 37-bar truss structure, as well as an experimental test of 8-story shear-type steel frame structure in the laboratory. Compared with BOA and DE, the numerical and experimental results show that the proposed HBODEA is more robust to detect the reduction of stiffness with limited sensors and contaminated measurements. In addition, the effect of search space, two dynamic operators, population size on identification accuracy and efficiency of the proposed identification strategy are further investigated.

• The Journal of Advanced Prosthodontics
• /
• v.4 no.4
• /
• pp.197-203
• /
• 2012

PURPOSE. The aim of this systematic review was to address treatment outcome according to attachment systems for mandibular implant overdentures in terms of implant survival rate, prosthetic maintenance and complications, and patient satisfaction. MATERIALS AND METHODS. A systematic literature search was conducted using PubMed and hand searching of relevant journals considering inclusion and exclusion criteria. Clinical trial studies on mandibular implant overdentures until August, 2010 were selected if more than one type of overdenture attachment was reported. Twenty four studies from 1098 studies were finally included and the data on implant survival rate, prosthetic maintenance and complications, patient satisfaction were analyzed relative to attachment systems. RESULTS. Four studies presented implant survival rates (95.8 - 97.5% for bar, 96.2 - 100% for ball, 91.7% for magnet) according to attachment system. Ten other studies presented an implant survival rate ranging from 93.3% to 100% without respect to the attachment groups. Common prosthetic maintenance and complications were replacement of an assay for magnet attachments, and activation of a matrix or clip for ball or bar attachments. Prosthetic maintenance and complications most commonly occurred in the magnet groups. Conflicting findings were found on the rate of prosthetic maintenance and complications comparing ball and bar attachments. Most studies showed no significant differences in patient satisfaction depending upon attachment systems. CONCLUSION. The implant survival rate of mandibular overdentures seemed to be high regardless attachment systems. The prosthetic maintenance and complications may be influenced by attachment systems. However patient satisfaction may be independent of the attachment system.

• The KIPS Transactions:PartD
• /
• v.8D no.3
• /
• pp.285-294
• /
• 2001

Recently many efforts on the development of automatic processing system for delivery sequency sorting have been performed in ETRI, which requires the use of postal 4-state bar code system to encode delivery points. This paper addresses the issue on the extension of read range and the improvement of image processing method. For the improvement of image processing procedure, applied information acquisition method through basic two thresholds onto the horizontal axial line of gray image based on reference information of 4-state bar code symbology. Symbol values are computed after creating two threshold values based on the obtained information through search of horizontal axial values. The implementation result of 4-state bar code reader are obtained the symbol values within 30~60 msec (58,000~116,000 mail item/hour)without noise removal or image rotation in spite of the incline $\pm 45^{\circ}$.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.12 no.4
• /
• pp.114-122
• /
• 2013

This study introduces an optimization methodology for the determination of gate location that ensures the melt flow balance within a part cavity of injection mold. A new sequential direct-search scheme based on the recursive reduction of the designer-specified gate design area is developed, and it is integrated with a commercial flow simulation tool for optimization. To quantify the level of melt flow balance, we employ the maximum difference among the fill times for the melt fronts to reach the boundary elements of part cavity as objective function. The proposed methodology is successfully applied in the case study of melt flow balancing in molding of a bar code scanner model. The result shows that the melt flow balance at the optimized gate positions is significantly improved from that for the initial gate position.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.21 no.3
• /
• pp.27-34
• /
• 2007

The rotor faults of induction motors may cause bad effects on the performance of the induction motor. This paper proposes the detecting technique of these faults by analyzing the waveform of the induced current and voltage of search coil using numerical analysis and the experiment. Several defective rotor bars are simulated to analyze the fault conditions-broken bars and high resistance of rotor bars. In order to prove the usefulness of the proposed method, we made an prototype experimental apparatus. The waveform of the induced voltages in search coil has the obvious characteristics and it is easy to differentiate the normal rotor from the abnormal one. The experimental results show that the proposed method is useful to detect the rotor fault conditions.

• Journal of Ocean Engineering and Technology
• /
• v.21 no.2 s.75
• /
• pp.75-80
• /
• 2007

The design variables for structural systems, in most practical designs, are chosen from a list of discrete values, which are commercially available sizing. This paper presents the application of Genetic Algorithms for determining the optimum design for two-dimensional structures with discrete and pseudocontinuous design variables. Genetic Algorithms are heuristic search algorithms and are effective tools for finding global solutions for discrete optimization. In this paper, Genetic Algorithms are used as the method of Elitism and penalty parameters, in order to improve fitness in the reproduction process. Examples in this paper include: 10 bar planar truss and 1 bay 8-story frame. Truss with discrete and pseudoucontinuous design variables and steel frame with W-sections are used for the design of discrete optimization.

• Journal of Mechanical Science and Technology
• /
• v.15 no.8
• /
• pp.1132-1142
• /
• 2001

The paper describes the construction of global function approximation models for use in design optimization via global search techniques such as genetic algorithms. Two different approximation methods referred to as evolutionary fuzzy modeling (EFM) and neuro-fuzzy modeling (NFM) are implemented in the context of global approximate optimization. EFM and NFM are based on soft computing paradigms utilizing fuzzy systems, neural networks and evolutionary computing techniques. Such approximation methods may have their promising characteristics in a case where the training data is not sufficiently provided or uncertain information may be included in design process. Fuzzy inference system is the central system for of identifying the input/output relationship in both methods. The paper introduces the general procedures including fuzzy rule generation, membership function selection and inference process for EFM and NFM, and presents their generalization capabilities in terms of a number of fuzzy rules and training data with application to a three-bar truss optimization.

• Journal of Korean Association for Spatial Structures
• /
• v.8 no.3
• /
• pp.75-82
• /
• 2008

In this study, the modified genetic algorithm, D-GA, is proposed. D-GA is a hybrid genetic algorithm combined a simple genetic algorithm and the local search algorithm using direction vectors. Also, two types of direction vectors, learning direction vector and random direction vector, are defined without the sensitivity analysis. The accuracy of D-GA is compared with that of simple genetic algorithm. It is demonstrated that the proposed approach can be an effective optimization technique through a minimum weight structural optimization of ten bar truss.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1057-1058
• /
• 2007

This paper proposes an alternative vibration detection method in a squirrel-cage induction motor using flux sensors. The air-gap flux will be changed when mechanical vibration occurs by bearing fault as well as broken rotor bar and air-gap eccentricity. For detecting those flux variations due to vibration, search coils are installed at stator slots. The induction motor with 380 [V], 7.5 [kW], 4 [Poles], 1,760 [rpm] ratings is used. Magnitudes and distortion of the induced voltage from flux sensors are used to discriminate faulted types. As a result, the flux sensor has been proven to be useful for vibration detection. It is compared to the result with vibration sensor as well.