• The Korean Journal of Applied Statistics
• /
• v.25 no.5
• /
• pp.829-835
• /
• 2012

When the input features are generated by factors in a classification problem, it is more meaningful to identify important factors, rather than individual features. The $F_{\infty}$-norm support vector machine(SVM) has been developed to perform automatic factor selection in classification. However, the $F_{\infty}$-norm SVM may suffer from estimation inefficiency and model selection inconsistency because it applies the same amount of shrinkage to each factor without assessing its relative importance. To overcome such a limitation, we propose the adaptive $F_{\infty}$-norm ($AF_{\infty}$-norm) SVM, which penalizes the empirical hinge loss by the sum of the adaptively weighted factor-wise $L_{\infty}$-norm penalty. The $AF_{\infty}$-norm SVM computes the weights by the 2-norm SVM estimator and can be formulated as a linear programming(LP) problem which is similar to the one of the $F_{\infty}$-norm SVM. The simulation studies show that the proposed $AF_{\infty}$-norm SVM improves upon the $F_{\infty}$-norm SVM in terms of classification accuracy and factor selection performance.

• Structural Engineering and Mechanics
• /
• v.83 no.3
• /
• pp.363-375
• /
• 2022

Flexural capacity prediction is a challenging problem for externally prestressed concrete beams (EPCBs) due to the unbonded phenomenon between the concrete beam and external tendons. Many prediction equations have been provided in previous research but typically ignored the differences in deformation mode between internal and external unbonded tendons. The availability of these equations for EPCBs is controversial due to the inconsistent deformation modes and ignored second-order effects. In this study, the deformation characteristics and collapse mechanism of EPCB are carefully considered, and the ultimate deflected shape curves are derived based on the simplified curvature distribution. With the compatible relation between external tendons and the concrete beam, the equations of tendon elongation and eccentricity loss at ultimate states are derived, and the geometric interpretation is clearly presented. Combined with the sectional equilibrium equations, a rational and simplified flexural capacity prediction method for EPCBs is proposed. The key parameter, plastic hinge length, is emphatically discussed and determined by the sensitivity analysis of 324 FE analysis results. With 94 collected laboratory-tested results, the effectiveness of the proposed method is confirmed, and comparisons with the previous formulas are made. The results show the better prediction accuracy of the proposed method for both stress increments and flexural capacity of EPCBs and the main reasons are discussed.

• Structural Engineering and Mechanics
• /
• v.83 no.4
• /
• pp.563-576
• /
• 2022

This study aimed to evaluate the progressive collapse potential of buildings designed using conventional design codes for the merchant occupancy classification and subjected to a sudden column failure. For this purpose, three reinforced concrete buildings having different story numbers were designed according to the seismic design recommendations of TSCB-2019. Later on, the buildings were analyzed using the GSA-2016 and UFC 4-023-03 to observe their progressive collapse responses. Three columns were removed independently in the structures from different locations. Nonlinear dynamic analysis method for the alternate path direct design approach was implemented for the design evaluation. The plasticity of the structural members was simulated by using nonlinear fiber hinges. The moment, axial, and shear force interaction on the hinges was considered by the Modified Compression Field Theory. Moreover, an existing experimental study investigating the progressive collapse behavior of reinforced concrete structures was used to observe the validation of nonlinear fiber hinges and the applied analysis methodology. The study results deduce that a limited local collapse disproportionately more extensive than the initial failure was experienced on the buildings designed according to TSCB-2019. The mercantile structures designed according to current seismic codes require additional direct design considerations to improve their progressive collapse resistance against the risk of a sudden column loss.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.6
• /
• pp.748-755
• /
• 2006

In this paper, a trade-off design and fabrication of IPMC(Ionic Polymer Metal Composites) as an actuator for a flapping device have been described. Experiments for the internal solvent loss of IPMCs have been conducted for various combinations of cation and solvent in order to find out the best combination of cation and solvent for minimal solvent loss and higher actuation force. From the experiments, it was found that IPMCs with heavy water as their solvent could operate longer. Relations between length/thickness and tip force of IPMCs were also quantitatively identified for the actuator design from the tip force measurement of 200, 400, 640, and $800{\mu}m$ thick IPMCs. All IPMCs thicker than $200{\mu}m$ were processed by casting $Nafion^{TM}$ solution. The shorter and thicker IPMCs tended to generate higher actuation force but lower actuation displacement. To improve surface conductivity and to minimize solvent evaporation due to electrically heated electrodes, gold was sputtered on both surfaces of the cast IPMCs by the Physical Vapor Deposition(PVD) process. For amplification of a short IPMC's small actuation displacement to a large flapping motion, a rack-and-pinion type hinge was used in the flapping device. An insect wing was attached to the IPMC flapping mechanism for its flapping test. In this test, the wing flapping device using the $800{\mu}m$ thick IPMC. could create around $10^{\circ}{\sim}85^{\circ}$ flapping angles and $0.5{\sim}15Hz$ flapping frequencies by applying $3{\sim|}4V$.

• Journal of the Korean Institute of Gas
• /
• v.4 no.1 s.9
• /
• pp.26-32
• /
• 2000

The fretting wear corrosion characteristics between the SM20C and the SM20C, the YBsC3 and the STC4H was experimented by using radical type friction experimental device under the corrosion environment of atmosphere, neutral solution, acid solution and chemical factors of the sea water. The affection of underground water that affect fretting wear corrosion of the SM20C which is moving specimen was more sensitive at the STC4H and more insensible at the YBsC3. The affection of underground water that affect fretting wear corrosion of the STC4H was less, but in the $0.5\%\;H_2SO_4$ and $0.5\%\;HNO_3$ solutions the fretting wear corrosion of the STC4H was more large. The fretting wear corrosion of the SM20C which is moving specimen in the underground water was less than in the $3.5\%\;NaCl$, $0.5\%\;H_2SO_4$ and $0.5\%\;HNO_3$ solutions. As time passed, the fretting wear corrosion is increased in the $HNO_3$ solution and dull in the $0.5\%\;H_2SO_4$ solution.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.5
• /
• pp.569-576
• /
• 2011

In this paper a new analysis procedure for evaluation of progressive collapse resisting capacity of a structure was proposed based on the nonlinear static analysis procedure. The proposed procedure produces analysis results identical to those obtained by the linear static analysis procedure specified in the GSA guidelines without iteration, therefore saving a lot of computation time and excluding the possibility of human errors during the procedure. To verify the validity of the proposed procedure, the two methods were applied to the analysis of a reinforced concrete moment frame and a steel braced frame subjected to loss of a first story column and the results were compared. According to the analysis results, the two methods produce identical results in the prediction of progressive collapse and the hinge formation. As iterative analysis is not required in the proposed method, significant amount of analysis time is saved in the proposed analysis procedure.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.53-65
• /
• 2019

Object tracking is one of important steps to achieve video-based surveillance systems. Object tracking is considered as an essential task similar to object detection and recognition. In order to perform object tracking, various machine learning methods (e.g., least-squares, perceptron and support vector machine) can be applied for different designs of tracking systems. In general, generative methods (e.g., principal component analysis) were utilized due to its simplicity and effectiveness. However, the generative methods were only focused on modeling the target object. Due to this limitation, discriminative methods (e.g., binary classification) were adopted to distinguish the target object and the background. Among the machine learning methods for binary classification, total error rate minimization can be used as one of successful machine learning methods for binary classification. The total error rate minimization can achieve a global minimum due to a quadratic approximation to a step function while other methods (e.g., support vector machine) seek local minima using nonlinear functions (e.g., hinge loss function). Due to this quadratic approximation, the total error rate minimization could obtain appropriate properties in solving optimization problems for binary classification. However, this total error rate minimization was based on a batch mode setting. The batch mode setting can be limited to several applications under offline learning. Due to limited computing resources, offline learning could not handle large scale data sets. Compared to offline learning, online learning can update its solution without storing all training samples in learning process. Due to increment of large scale data sets, online learning becomes one of essential properties for various applications. Since object tracking needs to handle data samples in real time, online learning based total error rate minimization methods are necessary to efficiently address object tracking problems. Due to the need of the online learning, an online learning based total error rate minimization method was developed. However, an approximately reweighted technique was developed. Although the approximation technique is utilized, this online version of the total error rate minimization could achieve good performances in biometric applications. However, this method is assumed that the total error rate minimization can be asymptotically achieved when only the number of training samples is infinite. Although there is the assumption to achieve the total error rate minimization, the approximation issue can continuously accumulate learning errors according to increment of training samples. Due to this reason, the approximated online learning solution can then lead a wrong solution. The wrong solution can make significant errors when it is applied to surveillance systems. In this paper, we propose an exactly reweighted technique to recursively update the solution of the total error rate minimization in online learning manner. Compared to the approximately reweighted online total error rate minimization, an exactly reweighted online total error rate minimization is achieved. The proposed exact online learning method based on the total error rate minimization is then applied to object tracking problems. In our object tracking system, particle filtering is adopted. In particle filtering, our observation model is consisted of both generative and discriminative methods to leverage the advantages between generative and discriminative properties. In our experiments, our proposed object tracking system achieves promising performances on 8 public video sequences over competing object tracking systems. The paired t-test is also reported to evaluate its quality of the results. Our proposed online learning method can be extended under the deep learning architecture which can cover the shallow and deep networks. Moreover, online learning methods, that need the exact reweighting process, can use our proposed reweighting technique. In addition to object tracking, the proposed online learning method can be easily applied to object detection and recognition. Therefore, our proposed methods can contribute to online learning community and object tracking, detection and recognition communities.