• 한국건설관리학회논문집
• /
• 제21권6호
• /
• pp.113-124
• /
• 2020

노후화된 공동주택을 리모델링을 통하여 성능개선을 하고자 하는 사회적 움직임이 포착되고 있다. 이를 위해 리모델링의 공사비 분석, 구조 분석, 정치제도적 검토등이 진행되어 리모델링을 활성화방안이 제시되고 있다. 그러나 현재 공동주택 리모델링의 공사비 분석방법이 연구상으로 제시되고 있으나, 실무적 활용가능성에서 한계점이 존재하는 상황이다. 구체적으로, 실무적으로 활용되기 위해서는 현재 진행되었거나 진행 중인 사례에는 적용가능하지만 향후 발생할 사례 또한 공사비분석에 활용되기에 분석방법의 지속 가능성이 결여되어 있다. 따라서 본 연구에서는 공동주택 리모델링 공사비견적의 지속가능성을 위해 자동화된 견적방법을 제시하고자 한다. 공사비견적의 지속가능성을 위해서 Deep-Learning을 견적절차에 도입하였다. 구체적으로, 공동주택 리모델링에서 발생할 수 있는 설계요소,공종, 공사비 상승계수의 관계를 자동적으로 찾는 방안을 제시하였다. 추가적으로 Artificial Neural Network기반 견적의 태생적인 한계점인 불확실성 미반영을 보완하고자 Monte Carlo Simulation을 견적절차에 산입하였다. 사례가 누적될수록 더 높은 정확도를 제시하기 위하여 견적결과를 기존의 누적데이터와 비교를 하여 더 높은 정확도를 산출하는 방안 또한 제시되었다. 본 연구에서 제안한 자동화된 개산견적의 지속가능성을 검토하고자 13개 사례의 학습절차와 추가 2개 사례의 누적절차를 거쳤다. 결과적으로 추가된 2개 프로젝트의 특성이 반영된 새로운 공사비견적절차가 자동적으로 제시되었다. 본 연구에서는 15개의 사례를 활용하여 개산견적의 방안을 활용하였으나, 사례가 누적되어 반영될 경우 본 연구의 기대효과는 더욱 높아질 것으로 사료된다.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제21권1호
• /
• pp.94-99
• /
• 2015

비모수 베이스 통계학, 확률적 표집에 기반한 추론 등이 기계학습의 주요 패러다임으로 등장하면서 디리슐레(Dirichlet) 분포는 최근 다양한 그래프 모형 곳곳에 등장하고 있다. 디리슐레 분포는 일변수 감마 분포를 벡터 분포로 확장한 형태의 하나이다. 본 논문에서는 감마 분포를 갖는 임의의 자연수 X를 K개의 자연수의 합으로 임의 분할 할 때 각 부분의 크기 비율을 디리슐레 분포에서 표집하는 방법을 제안한다. 일반적으로 디리슐레 분포는 연속적인 (K-1)-단체(simplex) 위에 정의 되지만 자연수로 분할하는 표본은 자연수라는 조건 때문에 단체 내부의 이산 그리드 점에만 정의된다. 본 논문에서는 단체 위의 그리드 상의 이웃 점들의 확률 분포로부터 마르코프연쇄 몬테 칼로(MCMC) 제안 분포를 정의하고 일련의 표본들의 마르코프 연쇄를 구현하는 알고리듬을 제안한다. 본 방법은 마르코프 모델, HMM 및 준-HMM 등에서 각 상태별 시간 지속 분포를 표현하는데 활용 가능하다. 나아가 최근 제안된 전역-지역(global-local) 상태지속 분포를 동시에 모형화하는 감마-디리슐레 HMM에도 응용가능하다.

• 대한기계학회논문집A
• /
• 제41권8호
• /
• pp.729-736
• /
• 2017

물리모델과 기계학습방법을 이용한 모델지원탐지확률(MAPOD, Model-assisted Probability of Detection) 실험계획법과 운용 중 결함이 발생한 부품을 사용하여 탐지확률을 측정하는 방법을 연구하였다. 검사방법은 와전류탐상검사를 적용하였고 검사대상은 볼트홀 표면에 존재하는 피로균열이다. 모델 지원탐지확률을 이용한 결과 실험요인이 큰 폭으로 감소하였다. 몬테카를로(Monte Carlo) 시뮬레이션을 이용하여 시편 균열길이 측정의 불확실성을 탐지확률에 반영함으로써 사용 중 결함품을 사용하여 비파괴검사정비사의 기량검증을 수행할 수 있었다.

• Advances in nano research
• /
• 제12권5호
• /
• pp.529-547
• /
• 2022

Graphene is one of the strongest, stiffest, and lightest nanoscale materials known to date, making it a potentially viable and attractive candidate for developing lightweight structural composites to prevent high-velocity ballistic impact, as commonly encountered in defense and space sectors. In-plane twist in bilayer graphene has recently revealed unprecedented electronic properties like superconductivity, which has now started attracting the attention for other multi-physical properties of such twisted structures. For example, the latest studies show that twisting can enhance the strength and stiffness of graphene by many folds, which in turn creates a strong rationale for their prospective exploitation in high-velocity impact. The present article investigates the ballistic performance of twisted bilayer graphene (tBLG) nanostructures. We have employed molecular dynamics (MD) simulations, augmented further by coupling gaussian process-based machine learning, for the nanoscale characterization of various tBLG structures with varying relative rotation angle (RRA). Spherical diamond impactors (with a diameter of 25Å) are enforced with high initial velocity (V_i) in the range of 1 km/s to 6.5 km/s to observe the ballistic performance of tBLG nanostructures. The specific penetration energy (E_p^*) of the impacted nanostructures and residual velocity (V_r) of the impactor are considered as the quantities of interest, wherein the effect of stochastic system parameters is computationally captured based on an efficient Gaussian process regression (GPR) based Monte Carlo simulation approach. A data-driven sensitivity analysis is carried out to quantify the relative importance of different critical system parameters. As an integral part of this study, we have deterministically investigated the resonant behaviour of graphene nanostructures, wherein the high-velocity impact is used as the initial actuation mechanism. The comprehensive dynamic investigation of bilayer graphene under the ballistic impact, as presented in this paper including the effect of twisting and random disorder for their prospective exploitation, would lead to the development of improved impact-resistant lightweight materials.

• Journal of Information Processing Systems
• /
• 제8권3호
• /
• pp.409-420
• /
• 2012

The decision-making by agents in games is commonly based on reinforcement learning. To improve the quality of agents, it is necessary to solve the problems of the time and state space that are required for learning. Such problems can be solved by Macro-Actions, which are defined and executed by a sequence of primitive actions. In this line of research, the learning time is reduced by cutting down the number of policy decisions by agents. Macro-Actions were originally defined as combinations of the same primitive actions. Based on studies that showed the generation of Macro-Actions by learning, Macro-Actions are now thought to consist of diverse kinds of primitive actions. However an enormous amount of learning time and state space are required to generate Macro-Actions. To resolve these issues, we can apply insights from studies on the learning of tasks through Programming by Demonstration (PbD) to generate Macro-Actions that reduce the learning time and state space. In this paper, we propose a method to define and execute Macro-Actions. Macro-Actions are learned from a human subject via PbD and a policy is learned by reinforcement learning. In an experiment, the proposed method was applied to a car simulation to verify the scalability of the proposed method. Data was collected from the driving control of a human subject, and then the Macro-Actions that are required for running a car were generated. Furthermore, the policy that is necessary for driving on a track was learned. The acquisition of Macro-Actions by PbD reduced the driving time by about 16% compared to the case in which Macro-Actions were directly defined by a human subject. In addition, the learning time was also reduced by a faster convergence of the optimum policies.

• 전자공학회논문지C
• /
• 제35C권2호
• /
• pp.56-64
• /
• 1998

Multi-layered perceptrons that are a nonlinear neural network model, have been widely used for various applications mainly thanks to good function approximation capability for nonlinear fuctions. However, for digital hardware implementation of the multi-layere perceptrons, the quantization scheme using "look-up tables (LUTs)" is commonly employed to handle nonlinear signmoid activation functions in the neworks, and thus requires large amount of storage to prevent unacceptable quantization errors. This paper is concerned with a new effective methodology for digital hardware implementation of multi-layered perceptrons, and proposes a "piecewise affine approximation" method in which input domain is divided into (small number of) sub-intervals and nonlinear sigmoid function is linearly approximated within each sub-interval. Using the proposed method, we develop an expression and an error backpropagation type learning algorithm for a multi-layered perceptron, and compare the performance with the quantization method through Monte Carlo simulations on XOR problems. Simulation results show that, in terms of learning convergece, the proposed method with a small number of sub-intervals significantly outperforms the quantization method with a very large storage requirement. We expect from these results that the proposed method can be utilized in digital system implementation to significantly reduce the storage requirement, quantization error, and learning time of the quantization method.quantization method.

• Journal of information and communication convergence engineering
• /
• 제14권4호
• /
• pp.258-267
• /
• 2016

Three-dimensional integrated circuits (3D ICs) experience die-to-die variations in addition to the already challenging within-die variations. This adds an additional design complexity and makes variation estimation and full-chip optimization even more challenging. In this paper, we show that the industry standard on-chip variation (AOCV) tables cannot be applied directly to 3D paths that are spanning multiple dies. We develop a new machine learning-based model and methodology for an accurate variation estimation of logic paths in 3D designs. Our model makes use of key parameters extracted from existing GDSII 3D IC design and sign-off simulation database. Thus, it requires no runtime overhead when compared to AOCV analysis while achieving an average accuracy of 90% in variation evaluation. By using our model in a full-chip variation-aware 3D IC physical design flow, we obtain up to 16% improvement in critical path delay under variations, which is verified with detailed Monte Carlo simulations.

• International Journal of Aeronautical and Space Sciences
• /
• 제11권4호
• /
• pp.326-337
• /
• 2010

This work addresses problems regarding trajectory planning for unmanned aerial vehicle sensors. Such sensors are used for taking measurements of large nonlinear systems. The sensor investigations presented here entails methods for improving estimations and predictions of large nonlinear systems. Thoroughly understanding the global system state typically requires probabilistic state estimation. Thus, in order to meet this requirement, the goal is to find trajectories such that the measurements along each trajectory minimize the expected error of the predicted state of the system. The considerable nonlinearity of the dynamics governing these systems necessitates the use of computationally costly Monte-Carlo estimation techniques, which are needed to update the state distribution over time. This computational burden renders planning to be infeasible since the search process must calculate the covariance of the posterior state estimate for each candidate path. To resolve this challenge, this work proposes to replace the computationally intensive numerical prediction process with an approximate covariance dynamics model learned using a nonlinear time-series regression. The use of autoregressive time-series featuring a regularized least squares algorithm facilitates the learning of accurate and efficient parametric models. The learned covariance dynamics are demonstrated to outperform other approximation strategies, such as linearization and partial ensemble propagation, when used for trajectory optimization, in terms of accuracy and speed, with examples of simplified weather forecasting.

• Structural Engineering and Mechanics
• /
• 제84권3호
• /
• pp.323-335
• /
• 2022

The main idea of the framework is to seamlessly combine a reasonably accurate and fast surrogate model with the importance sampling strategy. Developing a surrogate model for predicting structures' dynamic responses is challenging because it involves high-dimensional inputs and outputs. For this purpose, a novel surrogate model based on cutting-edge deep learning architectures specialized for capturing temporal relationships within time-series data, namely Long-Short term memory layer and Transformer layer, is designed. After being properly trained, the surrogate model could be utilized in place of the finite element method to evaluate structures' responses without requiring any specialized software. On the other hand, the importance sampling is adopted to reduce the number of calculations required when computing the failure probability by drawing more relevant samples near critical areas. Thanks to the portability of the trained surrogate model, one can integrate the latter with the Importance sampling in a straightforward fashion, forming an efficient framework called TTIS, which represents double advantages: less number of calculations is needed, and the computational time of each calculation is significantly reduced. The proposed approach's applicability and efficiency are demonstrated through three examples with increasing complexity, involving a 1D beam, a 2D frame, and a 3D building structure. The results show that compared to the conventional Monte Carlo simulation, the proposed method can provide highly similar reliability results with a reduction of up to four orders of magnitudes in time complexity.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2000년도 가을 학술발표논문집 Vol.27 No.2 (2)
• /
• pp.90-92
• /
• 2000

데이터마이닝 문제는 데이터를 그 속성들에 따라 분류하여 예측하는 것뿐만 아니라 분류된 속성들간의 연관성에 대해 잘 설명할 수 있어야 한다. 일반적으로 변수들간의 연관성을 잘 설명할 수 있으면서도 높은 예측력을 가지는 방법으로는 베이지안 네트웍 분류자(Bayesian network classifier)가 있다. 그러나 이것은 데이터 마이닝과 같은 대용량 데이터에서는 성능이 떨어지는 단점이 있다. 이에 이 논문에서는 최근 RBF 신경망이 입력변수 선정문제에 성공적으로 적용된 Reversible Jump Markov Chain Monte Carlo 방법을 이용하여 최적의 입력변수들만을 선택하여 베이지안 네트웍을 학습하는 Selective BN Augmented Naive-Bayes Classifier를 새로운 방안으로 제안하고 이를 실제 데이터마이닝 문제에 적용한 결과를 제시한다.