• 한국지진공학회:학술대회논문집
• /
• 한국지진공학회 2003년도 추계 학술발표회논문집
• /
• pp.295-302
• /
• 2003

A quantitative approach for the retrofit prioritization of bridges is developed based on the damage risk of seismic vulnerable components. In the developed approach, seismic damage risk is estimated in the probabilistic perspectives with an analytical bridge model, which can consider various phenomena found in the seismic behaviors of girder-type bridges and damage models of various vulnerable components. Based on the total cost due to failure of structural components, weighting factors are proposed. Finally, the ranking index and retrofit priority of bridges are estimated from the overall damage risk and weighting factors of bridges. As a result, the retrofit priority of four PSC girder bridges is evaluated by using the proposed approach. The vulnerable components in need of seismic retrofit are selected accordingly. From simulated results, the validity of the proposed approach is verified by comparison with the existing approach. In addition, the proposed approach is found to be appropriate in evaluating the priority of existing bridges.

• 한국음향학회지
• /
• 제30권6호
• /
• pp.324-329
• /
• 2011

본 논문에서는 조건 사후 최대 확률 (conditional maximum a posteriori, CMAP)과 음성 스펙트럼 변이 조건을 기반으로 한 새로운 음성 검출기 (voice activity detection, VAD)를 제안한다. 제안된 음성 검출기는 통계적 모델을 기반으로 한 우도비 테스트 (likelihood ratio test, LRT)의 문턱값을 결정하는데 조건 사후 최대 확률과 스펙트럼 변이의 상태 값을 조건부 확률로 부과한다. 제안된 알고리즘을 다양한 잡음 환경에서 기존의 CMAP 기반의 음성 검출기와 비교한 결과 전체적으로 향상된 성능을 보였으며 특히 SNR이 낮은 조건에서 향상 폭이 컸다.

• Structural Engineering and Mechanics
• /
• 제23권3호
• /
• pp.217-232
• /
• 2006

A crack model for the fracture in concrete based on meshless method is proposed in this paper. The cracks in concrete are classified into micro-cracks or macro-cracks respectively according to their widths, and different numerical approaches are adopted for them. The micro-cracks are represented with smeared crack approach whilst the macro-cracks are represented with discrete cracks that are made up with additional nodes and boundaries. The widely used meshless method, Element-free Galerkin method, is adopted instead of finite element method to model the concrete, so that the discrete crack approach is easier to be implemented with the convenience of arranging node distribution in the meshless method. Rotating-Crack-Model is proved to be preferred over Fixed-Crack-Model for the smeared cracks of this composite crack model due to its better performance on mesh bias. Numerical examples show that this composite crack model can take advantage of the positive characteristics in the smeared and discrete approaches, and overcome some of their disadvantages.

• Smart Structures and Systems
• /
• 제32권1호
• /
• pp.61-81
• /
• 2023

This study presents an ensemble learning based Bayesian model updating approach for structural damage diagnosis. In the developed framework, the structure is initially decomposed into a set of substructures. The autoregressive moving average (ARMAX) model is established first for structural damage localization based structural motion equation. The wavelet packet decomposition is utilized to extract the damage-sensitive node energy in different frequency bands for constructing structural surrogate models. Four methods, including Kriging predictor (KRG), radial basis function neural network (RBFNN), support vector regression (SVR), and multivariate adaptive regression splines (MARS), are selected as candidate structural surrogate models. These models are then resampled by bootstrapping and combined to obtain an ensemble model by probabilistic ensemble. Meanwhile, the maximum entropy principal is adopted to search for new design points for sample space updating, yielding a more robust ensemble model. Through the iterations, a framework of surrogate ensemble learning based model updating with high model construction efficiency and accuracy is proposed. The specificities of the method are discussed and investigated in a case study.

• Journal of Information Processing Systems
• /
• 제18권2호
• /
• pp.268-281
• /
• 2022

The development of information technology is bringing many changes to everyday life, and machine learning can be used as a technique to solve a wide range of real-world problems. Analysis and utilization of data are essential processes in applying machine learning to real-world problems. As a method of processing data in machine learning, we propose an approach based on applying multiple linear regression models by interlacing data to the task of classifying similar software. Linear regression is widely used in estimation problems to model the relationship between input and output data. In our approach, multiple linear regression models are generated by training on interlaced feature data. A combination of these multiple models is then used as the prediction model for classifying similar software. Experiments are performed to evaluate the proposed approach as compared to conventional linear regression, and the experimental results show that the proposed method classifies similar software more accurately than the conventional model. We anticipate the proposed approach to be applied to various kinds of classification problems to improve the accuracy of conventional linear regression.

• 한국수자원학회논문집
• /
• 제49권2호
• /
• pp.165-175
• /
• 2016

본 연구에서는 일사량에 기초한 증발량 산정방법의 적용성을 검토하기 위해 기존에 연구자들에 의해서 제안된 식들을 3가지 형태의 model group (Model groups A, B, and C)으로 분류하고, 이를 우리나라 6개 지역(서울, 대전, 전주, 부산, 목포, 제주)에 적용하였다. 증발접시 증발량 자료를 이용하여 이들 model group들의 매개변수를 추정하고, 검증하였다. 또한 Penman (1948) 조합식을 적용하여 이들 model group들과 비교하였다. 연구결과에 의하면 모든 지역에서 Nash-Sutcliffe (N-S) 효율지수가 0.663 이상을 보여서 만족스러운 증발량 산정결과를 보였다. 모형 검증과정에서 산정된 N-S 효율지수는 모든 연구지역에서 0.526이상을 보여서 역시 만족스러운 결과를 보였으나, 부산지역에서 적용된 Model groups B와 C를 제외하고는 모두 Penman (1948) 조합식보다 작은 N-S 효율지수를 보였다. 따라서 주요 기상자료 일부(풍속, 상대습도)가 부족하거나 측정되지 않는 경우에 증발량 산정을 위해서 Penman (1948) 조합식을 대체하여 일사량자료에 기초한 증발량 산정 방법이 적용될 수 있을 것으로 사료된다.

• 한국경영과학회지
• /
• 제23권3호
• /
• pp.123-133
• /
• 1998

A cell design model based on operation sequence is proposed for maximizing the total parts flow within cells considering the data of Process plans for parts, Production volume, and cell size. A relationship between machines is calculated on the basis of the process plans for parts obtained from process plan sheets. Then the machines are classified into machine cells using the relationship. The model is formulated as a 0-1 integer programming and a genetic algorithm approach is developed to solve the model. The developed approach is tested and Proved using actual industrial data. Experimental results indicate that the approach is appropriate for large-size cell design problems efficiently.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.1100-1102
• /
• 2000

Most systems in tile real world are non-linear and can be represented by the non-linear autoregressive moving average (NARMA) model. The extension of fuzzy system for modeling the system that is represented by NARMA model will be proposed in this paper. Here, fuzzy basis function (FBF) is used as fuzzy NARMA(p,q) model. Then, an approach to Identify fuzzy NARMA models based on fuzzy basis functions is proposed. The efficacy of the proposed approach is shown from experimental results.

• 한국컴퓨터정보학회논문지
• /
• 제23권9호
• /
• pp.43-50
• /
• 2018

In this paper, we propose a hexagon model-based efficient beacon frame scheduling approach for wireless sensor networks. The existing beacon frame scheduling approaches use a lot of slots and subslots for the beacon frame scheduling. Thus, the data from source nodes are not efficiently delivered to a sink node. Also in case a sink node needs to broadcast a beacon frame to the nodes in the network, delivering the beacon frame to the network nodes is not efficient as well. Thus, to solve the problem, we use a hexagon model to find the number of slots and subslots for the beacon frame scheduling. By using them for the beacon frame scheduling, the proposed approach performs better than other approaches in terms of the data transmission delay, the number of received data, the beacon transmission delay and the number of relaying the beacon frames.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.490-495
• /
• 1992

A neural model predictive control strategy combining a neural network for plant identification and a nonlinear programming algorithm for solving nonlinear control problems is proposed. A constrained nonlinear optimization approach using successive quadratic programming cooperates with neural identification network is used to generate the optimum control law for the complicate continuous/batch chemical reactor systems that have inherent nonlinear dynamics. Based on our approach, we developed a neural model predictive controller(NMPC) which shows excellent performances on nonlinear, model-plant mismatch cases of chemical reactor systems.