• International Journal of Highway Engineering
• /
• v.10 no.4
• /
• pp.213-224
• /
• 2008

Flexible pavement responses to vehicular loading, such as critical stresses and strains, in each pavement layer, could be predicted by the multilayered elastic analysis. However, multilayered elastic theory suffers from major drawbacks including spatial dimension of a numerical model, material properties considered in the analysis, boundary conditions, and ill-presentation of tire-pavement contact shape and stresses. To overcome these shortcomings, three-dimensional finite element (3D FE) models are developed and numerical analyses are conducted to calculate pavement responses to moving load in this study. This paper introduces a methodology for an effective 3D FE to simulate flexible pavement structure. It also discusses the mesh development and boundary condition analysis. Sensitivity analyses of flexible pavement response to loading are conducted. The infinite boundary conditions and time-dependent history of calculated pavement responses are considered in the analysis. This study found that the outcome of 3D FE implicit dynamic analysis of flexible pavement that utilizes appropriate boundary conditions, continuous moving load, viscoelastic hot-mix asphalt model is comparable to field measurements.

• Journal of Advanced Navigation Technology
• /
• v.27 no.5
• /
• pp.674-684
• /
• 2023

In this paper, a real-time prediction technique for gravity disturbances is proposed using a multi-layer perceptron (MLP) model. To select a suitable MLP model, 4 models with different network sizes were designed to compare the training accuracy and execution time. The MLP models were trained using the data of vehicle moving along the surface of the sea or land, including their positions and gravity disturbance. The gravity disturbances were calculated using the 2160^th degree and order EGM2008 with SHM. Among the models, MLP4 demonstrated the highest training accuracy. After training, the weights and biases of the 4 models were stored in the embedded computer of the INS to implement the MLP network. MLP4 was found to have the shortest execution time among the 4 models. These research results are expected to contribute to improving the navigation accuracy of INS through gravity disturbance compensation in the future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1183-1189
• /
• 2021

There have been recently an increasing number of people working out at home. However, many of them do not have face-to-face guidance from experts, so they cannot effectively correct their wrong pose. This may lead to strain and injury to those doing home training. To tackle this problem, this paper proposes a video data-based pose classification and correction system for home training. The proposed system classifies poses using the multi-layer perceptron and pose estimation model, and corrects poses based on joint angels estimated. A voting algorithm that considers the results of successive frames is applied to improve the performance of the pose classification model. Multi-layer perceptron model for post classification shows the highest accuracy with 0.9. In addition, it is shown that the proposed voting algorithm improves the accuracy to 0.93.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.5 s.39
• /
• pp.25-33
• /
• 2004

To evaluate the seismic capacity of a multistorey building structures in performance based seismic design, it is needed to convert MDOF model into equivalent SDOF model. This paper presents predictions for interstory drift of multistorey structures using method of converting a MDOF system into an equivalent SDOF model. The principal objective of this investigation is to evaluate appropriateness of converting method through performing nonlinear time history analysis of a multistory building structures and an equivalent SDOF model. Comparing the interstory drift of multistorey structures calculated by time history analysis and those evaluated by an equivalent SDOF model, the adequacy and the validity of converting method is verified. The conclusion of this study is following; A method of converting a MDOF system into an equivalent SDOF model through the nonlinear time history response analysis is valid. Inelastic first mode shapes are expected to be more accurate than elastic first mode shapes in obtaining interstory drift of multistorey structures from equivalent SDOF model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.130-132
• /
• 2001

신경망을 이용한 추천 기술은 항목이나 사용자간의 가중치를 학습할 수 있고, 자료 유형에 상관없이 데이터 처리가 용이하다. 또한 최근 연구를 통해서 그 우수성이 입증되고 있다. 그러나 사용자간의 상관관계로 추천하는 사용자 신경망 모델과 항목간의 상관관계로 추천하는 항목 신경망 모델이 서로 다른 관점으로 다른 선호도를 제시한 경우에 선택한 모델의 선호도에 따라 시스템의 성능이 좌우된다. 그러므로 효율적이고 성능이 우수한 추천 시스템을 위해 사용자와 항목 신경망 모델의 통합 방법을 제안한다. 두 모델 사이에 우선 순위를 결정하여 통합하는 순차적 통합 방법과 두 모델을 동시에 고려하는 병렬적 통합방법을 제안한다. 그러나 두 통합 방법은 선호도 예측 기준에 있어서 정적이고, 문제에 대한 적응성이 없다. 그러므로 신경망(퍼셉트론, 다층 퍼셉트론)을 이용한 통합 방법을 제안한다. 또한 퍼지의 소속함수를 이용하여 퍼지 추론를 적용한 통합 방법을 제안하고, 패턴 인식 분야에서 사용하는 BKS 방법을 적응하여 두 신경망 모델을 통합하여 실험한다. 본 논문에서는 사용자와 항목 신경망 모델을 통합함으로써 기존의 추천 기술인 연관 규칙과 단일 신경망 모델을 이용한 추천보다 우수함을 보이고 있다.

• Journal of the Korea Convergence Society
• /
• v.11 no.4
• /
• pp.33-39
• /
• 2020

Interest in 5G communication security has been growing recently amid growing expectations for 5G technology with faster speed and stability than LTE. However, 5G has so far included disparate areas, so it has not yet fully supported the issues of security. This paper proposes a blockchain-based IoT management model in order to efficiently provide the authentication of users using IoT in 5G In order to efficiently fuse the authentication of IoT users with probabilistic theory and physical structure, the proposed model uses two random keys in reverse direction at different layers so that two-way authentication is achieved by the managers of layers and layers. The proposed model applied blockchain between grouped IoT devices by assigning weights to layer information of IoT information after certification of IoT users in 5G environment is stratified on a probabilistic basis. In particular, the proposed model has better functions than the existing blockchain because it divides the IoT network into layered, multi-layered networks.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.713-719
• /
• 2009

In this paper, we develope a water demand forecasting algorithm using AR(Auto-regressive) and MLP(Multi-layer perceptron). To show effectiveness of the proposed method, we analyzed characteristics of time-series data collected in "A" purification plant at Jeon-Buk province during 2007-2008, and then performed the proposed method with various input factors selected through various analyses. As noted in experimental results, the performance of three types model such as multi-regressive, AR(Auto-regressive), and AR+MLP(Auto-regressive + Multi-layer perceptron) show 5.1%, 3.8%, and 3.6% with respect to MAPE(Mean Absolute Percentage Error), respectively. Thus, it is noted that the proposed method can be used to predict short-term water demand for the efficient operation of a water purification plant.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.916-919
• /
• 2018

심층신경망 모델은 우수한 성능을 갖고 있음에도 불구하고 모델이 어떤 판단 과정을 통해 결론을 내렸는지 파악하기 어렵다. 그에 따라 판단에 대한 근거가 중요한 분야에서는 심층신경망 모델을 적용한 실제 사례를 찾기 어렵다. 인공신경망 모델을 해석하기 어렵다는 문제를 해결하기 위해 내부 구조를 이용하여 규칙을 추출하는 decompositional 접근법이 제안되었으나 기존의 연구는 대부분 은닉층이 1개인 다층 퍼셉트론 모델에서 규칙을 생성하는 것을 가정하고 있다. 오늘날 사용하는 심층신경망 모델은 일반적으로 여러 은닉층을 가지고 있기 때문에 기존의 접근법을 그대로 적용할 경우 규칙 불확실성에 따라 잘못된 규칙을 추출하는 문제가 발생한다. 본 논문은 decompositional 접근법에 존재하는 규칙 불확실성 문제를 완화하고 깊이가 깊은 심층신경망 모델에 규칙을 추출하는 방법을 제안한다. 제안한 접근법은 실제 활성화 값을 통해 지식을 추출하며, 이를 통해 규칙 불확실성 문제를 완화할 수 있었다.

• Journal of the Korean earth science society
• /
• v.41 no.3
• /
• pp.291-306
• /
• 2020

In this study, we developed a learning progression for the structure of the solar system using multi-tier supply form items and validated its appropriateness. To this end, by applying Wilson's (2005) construct modeling approach, we set up 'solar system components,' 'size and distance pattern of solar system planets,' and 'solar system modeling' as the progress variables of the learning progression and constructed multi-tier supply form items for each of these variables. The items were applied to 150 fifth graders before and after the classes that dealt with the 'solar system and star' unit. To describe the results of the assessment, the students' responses to each item were categorized into five levels. By analyzing the Wright map that was created by applying the partial credit Rasch model, we validated the appropriateness of the learning progression based on the students' responses. In addition, the validity of the hypothetical pathway that was established in the learning progression was verified by tracking changes in the developmental level of students before and after the classes. The results of the research are as follows. The bottom-up research method that used multi-tier supply form items was able to elaborately set the empirical learning progression for the conceptualization of the structure of the solar system that is taught in elementary school. In addition, the validity of the learning progression was high, and the development of students was found to change with the learning progression.

• 한국노년학
• /
• v.39 no.4
• /
• pp.977-996
• /
• 2019

The current study aimed to examine the gender differences in trajectories of nine successful aging indicators (chronic disease, depression, activities of daily living, instrumental activities of daily living, mini mental state examination, social activity, personal contact, health satisfaction, and general life satisfaction) with age, controlling the effect of education. The data were from the Korean Longitudinal Study of Aging, which had been conducted biennially from 2006 to 2016. The sample included 822 men and 1,236 women who responded to all of the panel surveys and were 65 years old or above in 2006. Multilevel modeling analyses showed that older men had fewer chronic diseases; lower levels of depression; higher levels of activities of daily living, cognitive function, and social activity; and better perceived health satisfaction and general life satisfaction at age 65 years compared to women. However, both men and women showed increase in the number of chronic diseases and depression level, and decrease in physical, cognitive, and social functions with age. In addition, perceived health and life satisfaction also decreased after the age of 65. The trajectories of most of the indicators were non-linear, and markedly increased or decreased around mid-70s. Study limitations and implications were further discussed.