• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.24 no.4
• /
• pp.95-102
• /
• 2010

As the electric power industry is evolving into competitive market scheme, a new paradigm is required for the operation of market. Traditional dispatch algorithm was built based on the optimization model with an objective function and multiple constraints. Commercial market simulator followed the concept of the microeconomic model used in the dispatch algorithm, which is called as analytic method. On analytic method it is prerequisite to procure the exact data for the simulation. It is not easy anymore for each market participant to access to other participants' financial information while it used to be easy for monopoly decision maker to know all the information needed for the optimal operation. Considering the changing situation, it is required to introduce a new method for estimating the market price. This paper proposes an empirical method based on stochastic processes expected to build a capacity planning and long term contracts.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.6 no.4
• /
• pp.1-9
• /
• 1986

In this paper research results of developing a method of selecting design variables of an optimization problem from a finite set of pre-specified numbers, which can be utilized for the structural optimization with standardized structural members, is presented. The method first finds a continuous optimum under the assumption that design variables can be varied continuously. Then a pseudo-optimum is determined by selecting numbers from the set that are near to the continuous optimum and do not violate constraints. The pseudo-optimum is further improved to obtain the final discrete optimum from the set which minimizes cost function of the problem. In this research, the method is combined with the gradient projection optimization algorithm. The method is applied to several minimum weight truss optimization problems with constraints on the stresses, displacements, and design variables. As the results, it is found that the method can be efficiently applied to various optimization problems of which design variables must be chosen from a standard.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.247-247
• /
• 2011

도시유역에서의 일반적인 우수관거 설계 및 기존의 연구들은 수리학적 조건을 만족시키는 범위 내에서 최소의 비용을 위한 관거 설계에 초점을 맞추어왔다. 그러나 최적의 우수관거 설계는 우수관거의 근본적 목적인 내수배제를 최대화 하고자 하는 것이며, 이것은 관망 구성에 따른 시스템 내 유입 수문곡선들 간의 중첩효과를 제어함으로써 가능하다. 본 연구에서는 관내 흐름의 변화를 관망의구성에 따라서 제어될 수 있는 목적함수로서 다루고 있는 이정호(2010)의 우수관망 최적설계 모형에 대하여 각기 다른 유역 특성을 나타내는 실제 도시유역에서의 적용성을 검토하였다. 본 연구에서 이용한 최적 우수관망 노선 결정 모형은 이정호(2010)가 개발한 모형으로 도시유역에서의 우수관망의 노선 결정을 유출구 첨두유출량 최소화를 목적으로 최적화기법을 이용한 모형이다. 이정호(2010)는 개발된 모형을 통하여 관망 노선을 결정한 결과 설계빈도를 초과하는 강우에 대해서 내수침수 발생이 저감되는 효과를 가져올 수 있음을 분석하였다. 본 연구에서는 이러한 이정호(2010)의 모형에 대한 적용성을 검증하기 위하여 각기 다른 유역 특성을 갖는 실제 도시유역들에 대하여 해당 모형을 적용하였다. 이때 유역의 특성은 지표유출 및 첨두유출량에 지배적인 영향을 미치는 유역의 지표 경사에 대하여 고려되었다. 적용된 도시유역은 총 3개 배수분구로서 평균 지표 경사는 0.008, 0.006, 0.002로 각기 높은 경사, 보통 경사 및 완만한 경사 지대를 염두에 둔 유역 선정에 해당한다. 이에 따라 각 유역에 대하여 우수관망 노선을 최적화한 결과 설계빈도의 강우지속기간 30분 강우에 대하여 현재의 관망 대비 최적화된 관망에서의 첨두유출량 저감 비율은 지표경사 0.008인 유역에서는 약 7%, 지표경사 0.006인 유역에서는 약 17%, 지표경사 약 0.002인 유역에서는 약 8%로 나타났다. 따라서 본 연구결과 유역의 경사 특성에 따른 최적 우수관망을 통한 첨두유출량의 저감 비율간에는 상관관계를 정의할 수 없으며, 단지 우수관망에서의 노선 변경이 가능한 맨홀 지점이 어떤 중요한 지점에 얼마나 위치하느냐에 따라 달라지게 되는 것으로 분석되었다.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.447-453
• /
• 2020

In this paper, we propose a system to extract effective speaker representations from a speech signal using a deep learning method. Based on the fact that speech signal contains identity unrelated information such as text content, emotion, background noise, and so on, we perform a training such that the extracted features only represent speaker-related information but do not represent speaker-unrelated information. Specifically, we propose an auto-encoder based disentanglement method that outputs both speaker-related and speaker-unrelated embeddings using effective loss functions. To further improve the reconstruction performance in the decoding process, we also introduce a discriminator popularly used in Generative Adversarial Network (GAN) structure. Since improving the decoding capability is helpful for preserving speaker information and disentanglement, it results in the improvement of speaker verification performance. Experimental results demonstrate the effectiveness of our proposed method by improving Equal Error Rate (EER) on benchmark dataset, Voxceleb1.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.13 no.4
• /
• pp.417-425
• /
• 2000

This paper describes the application of discretized continuum-type optimality criteria(DCOC) and the development of optimum design program for the reinforced concrete continuous beams with rectangular cross-section. The cost of construction as objective function which includes the costs of concrete, reinforcing steel and formwork is minimized. The design constraints include limits on the maximum deflection, flexural and shear strengths, in addition to ductility requirements, and upper and lower bounds on design variables as stipulated by the design Code. Based on Kuhn-Tucker necessary conditions, the optimality criteria are explicitly derived in terms of the design variables-effective depth, and steel ratio. The self-weight of the beam is included in the equilibrium equation of the real system. An iterative procedure and computer program for updating the design variables are developed. Two numerical examples of reinforced concrete continuous beams are presented to show the applicability and efficiency of the DCOC-based technique.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.1
• /
• pp.48-55
• /
• 2016

The assessment and prediction of service life of a structure are usually under uncertainty so that rational probabilistic concepts and methods have to be applied. Based on these rational assessment and prediction, optimum maintenance strategies to minimize the life-cycle cost and/or maximize the structural safety can be established. The service life assessment and prediction considering maintenance actions generally includes effects of maintenance types and times of the structural components on the service life extensions of structural system. Existing researches on the service life management have revealed the appropriate system modeling considering the correlation among the components is required for system reliability analysis and probabilistic service life estimation. However, the study on correlation among the maintenance actions is still required. This paper deals with such a study for more effective and efficient service life management. In this paper, both the preventive and essential maintenances are considered for the extended service life estimation and management.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.10
• /
• pp.11-18
• /
• 1998

A binary line code, called MB46d, is designed by use of the BUDA(Binary Unit DSV and ASV) cell concept to retain the property of being runlength limited, DC tree, and with a power spectral null at the Nyquist frequency. This new code is a stateless line code with a simple encoding and a decoding rule and enables efficient error monitoring. The power spectrum and the eye pattern of the new line code are simulated for a minimum-bandwidth digital transmission system where the sinc function is used as a basic pulse. The obtained power null at the Nyquist frequency is wide enough to enable easy band-limiting as well as secure insertion of a clock pilot where necessary. The eye is also substantially wide to tolerate a fair amount of timing jitter in the receiver.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.15 no.4
• /
• pp.14-22
• /
• 2001

In this paper, we design the Adaptive Neuro-Fuzzy Precompensator(ANFP) for the suppression of low-frequency oscillation and the improvement of system stability. Here, ANFP is designed to compensate the conventional Power System Stabilizer(PSS). This design technique has the structural merit that is easily implemented by adding ANFP to an existing PSS. Firstly, the Fuzzy Precompensator with Loaming ability is constructed and is directly learned from the input and output data of the generating unit. Because the ANFP has the property of learning, fuzzy rules and membership functions of the compensator can be automatically tuned by teaming algorithm Loaming is based on the minimization of the ems evaluated by comparing the output of the ANFP and a desired controller. Case studies show the 7posed schema can be provided the good damping of the power system over the wide range of operating conditions and improved dynamic performance of the system.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.4
• /
• pp.36-44
• /
• 2012

Electrical impedance tomography is a relatively new imaging modality in which the internal conductivity (or resistivity) distribution of a object is reconstructed based on the injected currents and measured voltages through the electrodes placed on the surface of the object. In this paper, it is assumed that the relationship between the resistivity distribution and the resistance of electrodes is linear. From this linear relation, the weighting matrix can be obtained and modified iterative Landweber method is applied to estimate the internal resistivity distribution. Additionally, to accelerate the convergence rate and improve the spatial resolution of the reconstructed image, optimal step lengths for the iterative Landweber method are computed from the objective function in the least-square sense. The numerical experiments have been performed to illustrate the superior reconstruction performance of the proposed scheme.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.5B
• /
• pp.498-508
• /
• 2009

In wireless mobile environments, large bandwidth and high QoS (Quality of Service) are recently required to support the increased demands for multimedia services. LTE (Long Term Evolution) is one of the promising solutions for the next generation broadband wireless access systems. To efficiently use downlink resource and effectively support QoS, packet scheduling algorithm is one of the important features in LTE system. In this paper, we proposed DAPS (Delay-Aware Packet Scheduling) algorithm to consider QoS requirements of delays for various traffic classes as well as channel condition and fairness. To reflect delay experiences at scheduling instance, DAPS observes how queue waiting time of packet is closed to maximum allowable delay. The simulation results show that the DAPS algorithm yields better performance for delay experience by increasing the number of transmitted packets with satisfying the required delay time compared with existing scheduling algorithms.