Study on Different Representation Methods for Subspace Segmentation

Jiangshu Wei,Mantao Wang,Qianqian Wu 보안공학연구지원센터 2015 International Journal of Grid and Distributed Comp Vol.8 No.1

With many engineering and science application problems, we must deal with a lot of high-dimensional data, such as videos, images, web documents, text, etc. In the areas of computer vision, image processing and machine learning, high-dimensional data are widespread. However, it is very hard for obtaining meaningful learning and inference from these high-dimensional data directly, the computational complexity of high-dimensional data is often exponential. However, under many conditions, high-dimensional data lie in low-dimensional data corresponding to some classes of the data. Thus, finding the low-dimensional structure from the high-dimensional data is very important. The aim of subspace segmentation is to cluster data that lie in a union of low-dimensional subspaces. In recent years, based on the research of representation methods, many subspace segmentation algorithms appeared. Although these methods are all effective for handling subspace segmentation problems, they all have advantages and disadvantages. This paper focuses on the performance comparison of different subspace segmentation algorithms currently used in handling subspace segmentation problems and views other conventional methods that can be applied in this field.

A Tower-Shaped Three-Dimensional Piezoelectric Energy Harvester for Low-Level and Low-Frequency Vibration

Xiaoxiang Wei,Haibo Zhao,Junjie Yu,Yiming Zhong,Yanlin Liao,Shiwei Shi,Peihong Wang 한국정밀공학회 2021 International Journal of Precision Engineering and Vol.8 No.5

The multiple forms of vibration exist in an ambient environment diffusely and already become a considerable object for energy harvesting. However, how to effectively extract low-level, low-frequency, and multi-directional vibration from the ambient environment is becoming a key issue in the field of energy harvesting. To solve this issue, a tower-shaped piezoelectric vibration energy harvester (TS-PVEH) is reported. Finite element simulation indicates that TS-PVEH works in two fundamental modes, i.e., its in-plane and out-of-plane vibration modes. Meanwhile, simulation results show that the natural frequency of TS-PVEH is 3.39 Hz, 3.40 Hz, and 11.50 Hz, respectively; and the experiments also verified that. By virtue of the tower structure of TS-PVEH, the device is pretty sensitive to three-dimensional vibration. At a low level of acceleration 1 m/s 2 , the maximum load power of TS-PVEH is 65.8 μW in out-of-plane mode and 17.2 μW in in-plane mode, respectively. Furthermore, the effects of the PVDF connection mode on the output performance of TS-PVEH were studied in detail, and comparative experimental results show that a reasonable connection of PVDF can improve energy harvesting efficiency. The proposed TS-PVEH is expected to be used to scavenge energy from multi-dimensional, low-level, and low-frequency vibrations that present in an ambient environment.

Review of Low-Dimensional Nanomaterials for Blue-Light Emission

최원국 한국센서학회 2023 센서학회지 Vol.32 No.6

Low-dimensional (zero-dimensional (0-dim), 2-dimensional (2-dim)) nanoparticles, such as chalcogenide compound semiconductors,III-V semiconductors, transition metal dichalcogenides (TMDs), II-VI semiconductors, nanocarbons, hybrid quantum dots (QDs), andperovskite QDs (PQDs), for which blue light emission has been observed, are reviewed. Current synthesis and device fabrication technologiesas well as their prospective applications on next-generation quantum-dot-based light-emitting diodes are discussed.

저 차원 Hankel 구조를 이용한 PMU 데이터 복구에 관한 연구

신정훈,남수철,Evangelous Farantatos,Meng Wang,성태응 대한전기학회 2020 전기학회논문지 Vol.69 No.11

Phasor Measurement Units (PMUs) provide synchronized phasor measurements at much higher sampling rate than that in the traditional Supervisory Control And Data Acquisition (SCADA) system. Several synchrophasor-based algorithms and techniques have been and continue to be developed for real-time operation applications such as state estimation, stability analysis, disturbance detection, dynamic security assessment etc. However, synchrophasor data quality limits the incorporation of synchrophasor-based applications into control room operations environment and processes. The goal of this project is to develop methods that can improve synchrophasor data quality by recovering missing data reliably and efficiently. Data recovery refers to methods that estimate the values of missing data in the synchrophasor streams. Recently, modeless missing data recovery methods have been developed, that exploit the low-rank property of the spatial-temporal synchrophasor data blocks. A spatial-temporal synchrophasor data block can be considered as a matrix that is constructed by the measurements sampled at consecutive time instants with each row denoting the measurement of one certain channel across time. By exploiting the low-rank property of synchrophasor data matrices, the missing data recovery can be formulated as a low-rank matrix completion problem. The low-rank matrix completion problem has been extensively studied in the past few years and several algorithms have been developed to recover a low-rank matrix from partial observations. In this study, synchrophasor data analysis has been combined with low-rank matrix completion theory to develop a missing synchrophasor data recovery technique and tool.

저차원 선형 모델을 이용한 하이브리드 협력적 여과

고수정(Su-Jeong Ko) 한국정보과학회 2009 정보과학회논문지 : 소프트웨어 및 응용 Vol.36 No.10

협력적 여과는 특별한 아이템에 대한 사용자의 선호도를 예측하는 데 사용하는 기술이다. 이러한 협력적 여과 기술은 사용자 기반 접근 방식과 아이템 기반 접근 방식으로 구분할 수 있으며, 많은 상업적인 추천 시스템에서 광범위하게 사용되고 있다. 본 논문에서는 저차원 선형 모델을 사용하여 사용자 기반과 아이템 기반을 통합하는 하이브리드 협력적 여과 방법을 제안한다. 제안한 방법에서는 저차원 선형모델 중 비음수 행렬 분해(NMF)를 이용하여 기존의 협력적 여과 시스템의 문제점인 희박성과 대용량성의 문제점을 해결한다. 협력적 여과 시스템에서 NMF를 이용하는 방법은 사용자를 의미 관계로 표현할 때 유용하게 사용되나 사용자-아이템 행렬의 평가값에 따라 정확도가 낮아질 수 있으며, 모델 기반의 방법이기 때문에 계산 과정이 복잡하여 동적인 추천이 불가능하다는 단점을 갖는다. 이러한 단점을 보완하기 위하여 제안된 방법에서는 NMF에 의해 군집된 그룹을 대상으로 TF-IDF를 이용하여 그룹의 특징을 추출한다. 또한, 아이템 기반에서 아이템간의 유사도를 계산하기 위하여 상호정보량(mutual information)을 이용한다. 오프라인 상에서 훈련집합의 사용자를 군집시키고 그룹의 특징을 추출한 후, 온라인 상에서 추출한 그룹의 특징을 이용하여 새로운 사용자를 가장 최적의 그룹으로 분류함으로써 사용자를 분류하는 데 걸리는 시간을 단축시켜 동적인 추천을 가능하게 하며, 사용자 기반과 아이템 기반을 병합함으로써 기존의 방법보다 정확도를 높인다. Collaborative filtering is a technique used to predict whether a particular user will like a particular item. User-based or item-based collaborative techniques have been used extensively in many commercial recommender systems. In this paper, a hybrid collaborative filtering method that combines user-based and item-based methods using a low-dimensional linear model is proposed. The proposed method solves the problems of sparsity and a large database by using NMF among the low-dimensional linear models. In collaborative filtering systems the methods using the NMF are useful in expressing users as semantic relations. However, they are model-based methods and the process of computation is complex, so they can not recommend items dynamically. In order to complement the shortcomings, the proposed method clusters users into groups by using NMF and selects features of groups by using TF-IDF. Mutual information is then used to compute similarities between items. The proposed method clusters users into groups and extracts features of groups on offline and determines the most suitable group for an active user using the features of groups on online. Finally, the proposed method reduces the time required to classify an active user into a group and outperforms previous methods by combining user-based and item-based collaborative filtering methods.

Three-Dimensional Theoretical Study on Flow Characteristics of a Spiral-Channel Viscous Micropump

Donghyuk Kang,Koichi Nishibe,Kotaro Sato,Kazuhiko Yokota 한국유체기계학회 2019 International journal of fluid machinery and syste Vol.12 No.2

Technological developments have led to micropumps playing an ever-greater role at the heart of micro total analysis systems. Spiral-channel viscous micropumps have been studied theoretically, numerically, and experimentally. However, high-accuracy performance predictions have not been achieved over a wide range of operating conditions. The present study proposes three-dimensional and quasi-three-dimensional theoretical expressions for predicting the pressure performance of spiral-channel viscous micropumps in low-Reynolds-number environments. The theoretical analysis is validated through a series of comparisons with numerical simulation results for the pressure performance curves, velocity distributions, velocity gradient distributions, and channel internal pressure distributions. Furthermore, the influence of the channel aspect ratio on the performance characteristics is investigated.

Field-induced Staggered Moment Stabilization in Frustrated Quantum Magnets

Burkhard Schmidt,Mohammad Siahatgar,Peter Thalmeier 한국물리학회 2013 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.62 No.10

For low-dimensional frustrated quantum magnets, the dependence of the staggered moment on a magnetic field is nonmonotonic: For small and intermediate fields, quantum fluctuations are gradually suppressed, leading to an increase of the staggered moment as a function of the field strength. For large applied magnetic fields, the classically expected field dependence is recovered,namely a monotonous decrease with increasing field strength. The staggered moment is eventually suppressed when reaching the fully polarized state at the saturation field. The quantitative analysis of this behavior is an excellent tool to determine the frustration parameter of a magnetic compound. We have developed a general finite-size scaling scheme for numerical exact-diagonalization data of low-dimensional frustrated magnets, which we apply to the recently measured field dependence of the magnetic neutron scattering intensity of Cu(pz)2(ClO4)2 in the framework of the S = 1/2 twodimensional (2D) J1-J2 Heisenberg model. We also apply linear spin-wave theory to complement our numerical findings. Our results show that Cu(pz)2(ClO4)2 is a quasi-2D antiferromagnet with intermediate frustration J2/J1 = 0.2.

Multi-dimensional electrostatic plasma simulations using the particle-in-cell method for the low-temperature plasmas for materials processing

조영현,Cheon Cheongbin,Park Heesung,Hur Min Young,Lee Hae June 한국물리학회 2022 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.80 No.8

With the increasing requirement for the analysis of nonlinear and transient behaviors in plasma systems, the demand for particle-in-cell (PIC) simulations is increasing. However, most plasma simulations have utilized fluid models because of the high computational cost of the PIC simulation. This article introduces the most recent advances in the electrostatic particlein- cell simulations of low-temperature plasmas for materials processing. The parallelization in multi-dimensional geometry using heterogeneous computing and the method to treat curved boundaries are explained to improve the PIC simulation. Test examples of two-dimensional hollow cathode discharges and three-dimensional magnetron sputtering systems are presented to investigate the kinetic effects of the system.

1차원 무기 반도체 신 물질 재료의 연구 개발 동향

류학기 한국세라믹학회 2018 세라미스트 Vol.21 No.2

In order to overcome the problems of existing low-dimensional materials (carbon nanotubes, graphene, transition metal dichalcogenides, etc) researches on new 1D materials have been studied. In the case of LiMo 3 Se 3 and Mo 6 S 9- x I x , continuous researches have been carried out for 3D bulk synthesis and atomic scale dispersion. Recently, quantum confinement effect of LiMo 3 Se 3 and bio-stability of Mo 6 S 9-x I x have been proven and various applications have started to be studied. In addition, device application results using new 1D materials such as Sb 2 Se 3 (optoelectronic devices using the property of effectively reducing exciton decay due to no dangling bond) and VS 4 (electrochemical energy storage using the space between 1-D nanostructures) have been reported very importantly. Therefore, it can be claimed that it has reached a very important time to find and synthesize new 1D materials and to report various characteristics not existing.

품질공학 : 소량 생산을 위한 수치적 측면의 통계적 공정관리

진창호 ( Chang Ho Chin ) 대한산업공학회 2005 대한산업공학회 추계학술대회논문집 Vol.2005 No.-