• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.527-531
• /
• 2008

For statistical modeling, the model parameters are usually estimated by maximizing a probability measure, such as the likelihood or the posterior. In contrast, a variational Bayesian method treats the parameters of a model as probability distributions and computes optimal distributions for them rather than values. It has been shown that this approach effectively avoids the overfitting problem, which is common with other parameter optimization methods. This paper applies a variational Bayesian technique to surface fitting for height field data. Then, we propose point cloud denoising based on the basic surface fitting technique. Validation experiments and further tests with scan data verify the robustness of the proposed method.

• Tribology and Lubricants
• /
• v.25 no.6
• /
• pp.399-403
• /
• 2009

The objective of this paper is to predict pitting initiation by using a contact analysis and subsurface stress. Contact stresses are obtained by contact analysis of a semi-infinite solid based on the use of influence functions. Subsurface stress field is obtained using rectangular patch solutions. It is used Mesoscopic multiaxial fatigue criterion to predict contact fatigue life. It is important to predict pitting initiation to enhance reliability of the mechanical elements. Pitting life prediction in the spur gears which are fundamental mechanical element is presented in this paper.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2009.06a
• /
• pp.173-174
• /
• 2009

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.1
• /
• pp.16-22
• /
• 2012

In this paper, we proposed a method of baseline correction for analysis of Raman spectra of platelets from Alzheimer's disease (AD) transgenic mice. Measured Raman spectra include the meaningful information and unnecessary noise which is composed of baseline and additive noise. The Raman spectrum is divided into the local region including several peaks and the spectrum of the region is modeled by curve fitting using Gaussian model. The additive noise is clearly removed from the process of replacing the original spectrum with the fitted model. The baseline correction after interpolating the local minima of the fitted model with linear, piecewise cubic Hermite and cubic spline algorithm. The baseline corrected models extract the feature with principal component analysis (PCA). The classification result of support vector machine (SVM) and maximum $a$ posteriori probability (MAP) using linear interpolation method showed the good performance about overall number of principal components, especially SVM gave the best performance which is about 97.3% true classification average rate in case of piecewise cubic Hermite algorithm and 5 principal components. In addition, it confirmed that the proposed baseline correction method compared with the previous research result could be effectively applied in the analysis of the Raman spectra of platelet.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.3
• /
• pp.91-99
• /
• 2024

Piping systems are crucial facilities used in various industries, particularly in areas related to daily life and safety. Piping systems are fixed to the main structures of buildings and facilities but do not support external loads and serve as non-structural elements performing specific functions. Piping systems are affected by relative displacements owing to phase differences arising from different behaviors between two support points under seismic loads; this can cause damage owing to the displacement-dominant cyclic behavior. Fittings and joints in piping systems are representative elements that are vulnerable to seismic loads. To evaluate the seismic performance and limit states of fittings and joints in piping systems, a high-stroke actuator is required to simulate relative displacements. However, this is challenging because only few facilities can conduct these experiments. Therefore, element-level experiments are required to evaluate the seismic performance and limit states of piping systems connected by fittings and joints. This study proposed a method to evaluate the seismic performance of an elbow specimen that includes fittings and joints that are vulnerable to seismic loads in vertical piping systems. The elbow specimen was created by connecting straight pipes to both ends of a 90° pipe elbow using flexible groove joints. The seismic performance of the elbow specimen was evaluated using a cyclic loading protocol based on deformation angles. To determine the margin of the evaluated seismic performance, the limit states were assessed by applying cyclic loading with a constant amplitude.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.05a
• /
• pp.1281-1282
• /
• 2010

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.464-466
• /
• 2019

CNN 은 합성곱 연산을 사용하는 인공신경망의 한 종류이다. 이러한 인공 신경망에서는 훈련 데이터에 대한 과도한 학습으로 인해 시험 데이터에 제대로 반응하지 못하는 오버피팅이 발생할 우려가 있다. 이를 해결하기 위해 DropOut 과 DropConnect 를 사용할 수 있다. 본 논문에서는 DropOut 과 DropConnect 를 통한 학습 정도를 실험을 통해서 비교해보고, 인공 신경망에서 이 방법의 효과를 살펴본다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.1395-1398
• /
• 2010

최근 GWAS(Genome-wide association study)로 인해 수십만 개의 SNP들이 사용 가능하게 되었다. 그러나 SNP 정보의 양이 방대하여 모든 SNP 조합을 검토하는 방식은 계산 비용이 클 뿐 아니라 오버피팅의 위험이 따른다. 본 논문에서는 필터링 기반 알고리즘인 SNPHarvester의 속도를 개선하고 평가함수를 상호정보량으로 대체하여 실험한다. 기존 SNPHarvester와 비교해 속도면에서 50%가 향상되었고 평가함수 면에서는 기존 SNPHarvester와 동일한 성능을 보였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.2
• /
• pp.57-71
• /
• 1996

차량 모델 비선형성의 주된 요인중 하나는 타이어의 비선형성이라고 할 수 있다. 타이어 모델도 간편화하기 위해 선형화된 타이어 모델을 적용할 경우에 저속 주행 또는 고속 주행이라고도 조향각이 적을 때는 문제가 없지만, 급격한 가감속과 과도한 조향각을 주었을 때는 타이어 미끄럼 각(Tire Slip Angle)이 급격히 변화되므로 선형화 된 타이어 모텔을 적용하지 못하게 된다. 그러므로 타이어와 지면 사이의 물리적 현상을 자세히 표현할 수 있는 비선형 타이어 모델을 적용하지 못하게 된다. 그러므로 타이어와 지면 사이의 물리적 현상을 자세히 표현할 수 있는 비선형 타이어 모델이 요구되어진다. 실험적 모델은 실제 차량의 실험 데이터를 바탕으로 커브 피팅(Curve Fitting)하여 타이어의 동특성을 표현하도록 모델링 하므로서 모델의 정확도를 높일 수 있는 반면 요구하는 계수들이 많아지게 되어 계산량이 증가되는 단점이 있다. 기존의 타이어 모델 연구 결과에 대해 분석하고, 관측 자료들을 바탕으로 FMFNN(Fuzzy Membership Function based Neural Network)을 이용한 함수 근사화로서 타이어 횡축력과 종축력의 모델링 방법을 제안하였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.65-70
• /
• 2022

As fast storage has become popular, the memory management system designed for hard disks needs to be reconsidered. In this paper, we observe that memory access latency is sensitive to the page size when fast storage is adopted. We find the reason from the TLB miss rate, which has the increased impact on the memory access latency in comparison with the page fault rate, and there is trade-off between the TLB miss rate and the page fault rate as the page size is varied. To handle such situations, we model the page fault rate and the TLB miss rate accurately as a function of the page size. Specifically, we show that the power fit and the exponential fit with two terms are appropriate for fitting the TLB miss rate and the page fault rate, respectively. We validate the effectiveness of our model by comparing the estimated values from the model and real values.