• Journal of KIISE
• /
• v.45 no.1
• /
• pp.15-21
• /
• 2018

Low-Rank Representation (LRR) based methods are widely used in many practical applications, such as face clustering and object detection, because they can guarantee high prediction accuracy when used to constructing graphs in graph - based semi-supervised learning. However, in order to solve the LRR problem, it is necessary to perform singular value decomposition on the square matrix of the number of data points for each iteration of the algorithm; hence the calculation is inefficient. To solve this problem, we propose an improved and faster LRR method based on the recently published Fast LRR (FaLRR) and suggests ways to introduce and optimize additional constraints on the underlying optimization goals in order to address the fact that the FaLRR is fast but actually poor in classification problems. Our experiments confirm that the proposed method finds a better solution than LRR does. We also propose Fast MLRR (FaMLRR), which shows better results when the goal of minimizing is added.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1956-1957
• /
• 2011

본 연구는 White LED를 이용하여 주변 밝기 변화에 빠르게 적응하는 퍼지 뉴로 Dimming Control System을 설계한다. 본 논문에서는 방사형기저함수 신경회로망(Radial Basis Function Neural Network: RBFNN)을 설계하여 실제 White LED Dimming Control System에 적용시켜 모델의 근사화 및 일반화 성능을 평가한다. 제안한 모델에서의 은닉층은 방사형기저함수를 사용하여 적합도를 구현하였고, 후반부의 연결가중치는 경사하강법을 사용한다. 이때 멤버쉽 함수의 중심점은 HCM 클러스터링 (Hard C-Means Clustering)을 적용하여 결정한다. 연결가중치는 4가지 형태의 다항식을 대입하여 출력을 평가하였다. 최종 출력의 최적화를 위하여 PSO(Particle Swarm Optimization)을 이용하여 은닉층 노드수 및 다항식 형태를 결정한다. 본 논문에서 제안한 LED Dimming Control System은 Atmega8535를 사용하여 PWM 제어 방식을 사용하고, 조도계(Cds)를 이용하여 LED의 밝기에 따른 주변의 밝기를 감지하여 조명에 적응시키는 방법을 적용하였다.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1644_1645
• /
• 2009

본 논문은 비정상 시계열 예측을 위한 다중모델 퍼지 시스템과, 제안된 시스템의 최적화를 위한 유전 알고리즘의 응용을 다룬다. 일반적으로, 퍼지 예측시스템의 성능은 비선형 데이터가 가지고 있는 다양한 패턴이나 법칙성, 경향 등을 잘 분석하고 시스템에 반영함으로써 개선될 수 있다. 따라서, 본 논문은 원형 시계열의 특성을 보다 잘 반영할 수 있는 그들의 차분데이터를 시스템에 적용하며, 생성 가능한 차분 데이터들 중 원형 시계열의 특징에 가까운 일부를 추출하여 다중모델 퍼지 예측 시스템을 구현함으로써 다양한 원형시계열의 패턴이나 법칙성 등이 고려될 수 있도록 하였다. 다중 모델 퍼지 시스템의 각각의 예측기에는 구조가 간단한 k-means 클러스터링 기법을 적용하여 구현의 용이성을 꽤하였으며, 성능평가를 통해 선택된 최종 예측기는 RCGKA(real-coded genetic k-means clustering algorithms)를 통해 더욱 최적화된 규칙기반을 가지게 함으로써 예측성능이 개선될 수 있도록 하였다. 본 논문에 사용된 최적화 기법인 RCGKA에는 또한 성능이 우수한 다양한 유전연산자를 도입하여 더욱 예측기 성능이 강화될 수 있도록 하였으며, 시뮬레이션을 통해 제안된 예측시스템의 효용성을 증명하였다.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1871_1872
• /
• 2009

본 논문 에서는 Type-2 퍼지 논리 시스템을 설계하고, 불확실한 정보를 갖는 입력 데이터에 대하여 Type-1 퍼지 논리 시스템과 성능을 비교한다. Type-1 퍼지 논리 시스템은 외부 잡음에 민감한 단점을 가지고 있는 반면, Type-2 퍼지 논리 시스템은 불확실한 정보를 잘 표현 할 수 있다. 따라서 Type-2 퍼지 논리 시스템을 이용하여 이러한 단점을 극복하고자 2가지의 모델을 설계한다. 첫 번째 모델은 규칙의 전 후반부가 Type-1 퍼지 집합으로 구성된 Type-1 퍼지 논리 시스템을 설계 한다. 두 번째는 규칙 전 후반부에 Type-2 퍼지 집합으로 구성된 Type-2 퍼지 논리 시스템을 설계한다. 여기서 규칙 전반부의 입력 공간 분할 및 FOU(Footprint Of Uncertainty)형성에는 FCM(Fuzzy C_Means) clustering 방법을 사용하고, 입자 군집 최적화(Particle Swarm Optimization) 알고리즘을 사용하여 최적의 파라미터를 설계한다. 본 논문 에서는 또한 입력 데이터에 인위적으로 가하는 노이즈에 따른 각각 모델의 성능을 비교한다. 마지막으로 비선형 모델 평가에 주로 사용되는 NOx 데이터를 제안된 모델에 적용하고, 실험을 통하여 노이즈가 첨가되고, 불확실한 정보를 다루기에 Type-1 퍼지 논리 시스템 보다 Type-2 퍼지 논리 시스템이 효율적이라는 것을 보인다.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2004.11a
• /
• pp.107-116
• /
• 2004

The advent of microarray technologies gives an opportunity to moni tor the expression of ten thousands of genes, simultaneously. Such microarray data can be deteriorated by experimental errors and image artifacts, which generate non-negligible outliers that are estimated by 15% of typical microarray data. Thus, it is an important issue to detect and correct the se faulty probes prior to high-level data analysis such as classification or clustering. In this paper, we propose a systematic procedure for the detection of faulty probes and its proper correction in Genechip array based on multivariate statistical approaches. Principal component analysis (PCA), one of the most widely used multivariate statistical approaches, has been applied to construct a statistical correlation model with 20 pairs of probes for each gene. And, the faulty probes are identified by inspecting the squared prediction error (SPE) of each probe from the PCA model. Then, the outlying probes are reconstructed by the iterative optimization approach minimizing SPE. We used the public data presented from the gene chip project of human fibroblast cell. Through the application study, the proposed approach showed good performance for probe correction without removing faulty probes, which may be desirable in the viewpoint of the maximum use of data information.

• Nuclear Engineering and Technology
• /
• v.52 no.10
• /
• pp.2262-2273
• /
• 2020

Digital twin technology can provide significant value for the prognostics and health management (PHM) of critical plant components by improving insight into system design and operating conditions. Digital twinning of systems can be utilized for anomaly detection, diagnosis and the estimation of the system's remaining useful life in order to optimize operations and maintenance processes in a nuclear plant. In this regard, a conceptual framework for the application of digital twin technology for the prognosis of Control Element Drive Mechanism (CEDM), and a data-driven approach to anomaly detection using coil current profile are presented in this study. Health management of plant components can capitalize on the data and signals that are already recorded as part of the monitored parameters of the plant's instrumentation and control systems. This work is focused on the development of machine learning algorithm and workflow for the analysis of the CEDM using the recorded coil current data. The workflow involves features extraction from the coil-current profile and consequently performing both clustering and classification algorithms. This approach provides an opportunity for health monitoring in support of condition-based predictive maintenance optimization and in the development of the CEDM digital twin model for improved plant safety and availability.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.1
• /
• pp.68-77
• /
• 2018

Regarding demand response (DR) by residential users (R-users), the users try to reduce electricity costs by adjusting their power consumption in response to the time-varying price. However, their power consumption may be affected not only by the price, but also by user convenience for using appliances. This paper proposes a methodology for appliance scheduling (AS) that considers the user convenience based on historical data. The usage pattern for appliances is first modeled applying the copula function or clustering method to evaluate user convenience. As the modeling results, the comfort distribution or representative scenarios are obtained, and then used to formulate a discomfort index (DI) to assess the degree of the user convenience. An AS optimization problem is formulated in terms of cost and DI. In the case study, various AS tasks are performed depending on the weights for cost and DI. The results show that user convenience has significant impacts on AS. The proposed methodology can contribute to induce more DR participation from R-users by reflecting properly user convenience to AS problem.

• Journal of Sensor Science and Technology
• /
• v.25 no.6
• /
• pp.431-434
• /
• 2016

Currently, the radial basis function network (RBFN) and various other neural networks are employed to classify gases using chemical sensors arrays, and their performance is steadily improving. In particular, the identification performance of the RBFN algorithm is being improved by optimizing parameters such as the center, width, and weight, and improved algorithms such as the radial basis function network-stochastic gradient (RBFN-SG) and radial basis function network-normalized stochastic gradient (RBFN-NSG) have been announced. In this study, we optimized the number of centers, which is one of the parameters of the RBFN-NSG algorithm, and observed the change in the identification performance. For the experiment, repeated measurement data of 8 samples were used, and the elbow method was applied to determine the optimal number of centers for each sample of input data. The experiment was carried out in two cases(the only one center per sample and the optimal number of centers obtained by elbow method), and the experimental results were compared using the mean square error (MSE). From the results of the experiments, we observed that the case having an optimal number of centers, obtained using the elbow method, showed a better identification performance than that without any optimization.

• Proceedings of the Korean Society of Computational and Applied Mathematics Conference
• /
• 2003.09a
• /
• pp.1-1
• /
• 2003

In this talk, a reduced-order modeling methodology based on centroidal Voronoi tessellations (CVT's)is introduced. CVT's are special Voronoi tessellations for which the generators of the Voronoi diagram are also the centers of mass (means) of the corresponding Voronoi cells. The discrete data sets, CVT's are closely related to the h-means clustering techniques. Even with the use of good mesh generators, discretization schemes, and solution algorithms, the computational simulation of complex, turbulent, or chaotic systems still remains a formidable endeavor. For example, typical finite element codes may require many thousands of degrees of freedom for the accurate simulation of fluid flows. The situation is even worse for optimization problems for which multiple solutions of the complex state system are usually required or in feedback control problems for which real-time solutions of the complex state system are needed. There hava been many studies devoted to the development, testing, and use of reduced-order models for complex systems such as unsteady fluid flows. The types of reduced-ordered models that we study are those attempt to determine accurate approximate solutions of a complex system using very few degrees of freedom. To do so, such models have to use basis functions that are in some way intimately connected to the problem being approximated. Once a very low-dimensional reduced basis has been determined, one can employ it to solve the complex system by applying, e.g., a Galerkin method. In general, reduced bases are globally supported so that the discrete systems are dense; however, if the reduced basis is of very low dimension, one does not care about the lack of sparsity in the discrete system. A discussion of reduced-ordering modeling for complex systems such as fluid flows is given to provide a context for the application of reduced-order bases. Then, detailed descriptions of CVT-based reduced-order bases and how they can be constructed of complex systems are given. Subsequently, some concrete incompressible flow examples are used to illustrate the construction and use of CVT-based reduced-order bases. The CVT-based reduced-order modeling methodology is shown to be effective for these examples and is also shown to be inexpensive to apply compared to other reduced-order methods.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.3
• /
• pp.343-349
• /
• 2009

In this paper, it would be shown that individuals can have different responses to the same email based on their preferences through computing the distributions of user clusters' email responses from clustering results based on email users' preference information. This paper presents an approach that incorporates user preferences to construct an anti-spam mail system, which is different from the conventional content-based ones. We consider email category information derived from the email content as well as user preference information. We also build a user preference ontology to formally represent the important concepts and rules derived from a data mining process and then apply a rule optimization procedure to exclude unnecessary rules. Experimental results show that our user preference based system achieves good performance in terms of accuracy, the rules derived from the system and human comprehensibility.