• Smart Structures and Systems
• /
• 제32권5호
• /
• pp.319-338
• /
• 2023

The study presents a new hybrid data-driven method by combining radial basis functions neural networks (RBF-NN) with the Jaya algorithm (JA) to provide effective structural health monitoring of arch dams. The novelty of this approach lies in that only one user-defined parameter is required and thus can increase its effectiveness and efficiency, as compared to other machine learning techniques that often require processing a large amount of training and testing model parameters and hyper-parameters, with high time-consuming. This approach seeks rapid damage detection in arch dams under dynamic conditions, to prevent potential disasters, by utilizing the RBF-NNN to seamlessly integrate the dynamic elastic modulus (DEM) and modal parameters (such as natural frequency and mode shape) as damage indicators. To determine the dynamic characteristics of the arch dam, the JA sequentially optimizes an objective function rooted in vibration-based data sets. Two case studies of hyperbolic concrete arch dams were carefully designed using finite element simulation to demonstrate the effectiveness of the RBF-NN model, in conjunction with the Jaya algorithm. The testing results demonstrated that the proposed methods could exhibit significant computational time-savings, while effectively detecting damage in arch dam structures with complex nonlinearities. Furthermore, despite training data contaminated with a high level of noise, the RBF-NN and JA fusion remained the robustness, with high accuracy.

• 대한교통학회지
• /
• 제22권6호
• /
• pp.101-108
• /
• 2004

서울시 대중교통개편에서 요금부과방안은 기본적으로 거리비례제체제(Distance-Based Fare System)에 근거하고 있다. 거리비례제에서 요금은 일정거리를 주행하는 기본요금과 수단간 환승에서 발생하는 환승요금, 일정거리 이상의 주행에 따른 할증요금으로 구분된다. 본 연구는 거리비례제에 따른 요금부과 시 최소요금경로를 탐색하는 방안을 제시한다. 이를 위한 다수의 수단이 존재하는 복합교통망의 환승지점에서 네트워크확장이 필요치 않도록 링크표지을 적용했다. 동일링크에서 복수통행수단의 표현이 가능하도록 수단에 따른 링크확장개념을 활용하였다. 따라서 본 연구에서는 제안하는 최소요금경로 알고리즘은 수단을 표현하기 위한 표식이 별도로 필요하지 않아, 기존의 링크표지 최적경로알고리즘의 적용이 가능하다. 또한 요금부과과정을 네트워크에 적용하기 위하여 출발지를 기준으로 표현된 연속된 두 링크에 대해 기본요금, 환승요금, 할증요금의 부과과정을 수식으로 표현하였다. 이 수식을 재귀(recursive)형태의 수식으로 전환하여 최소요금경로 탐색알고리즘을 제시하였다. 간단한 예제를 통하여 알고리즘 수행과정을 평가하였다.

• 대한임베디드공학회논문지
• /
• 제5권4호
• /
• pp.234-242
• /
• 2010

Two-way ranging methods such as TWR and SDS-TWR have been considered for many ranging systems because these methods are useful in the absence of synchronization. To estimate the location of a mobile node, complicated ranging procedures consisting of ranging frames between an anchor node and the mobile node are performed. Supporting multiple mobile nodes such as a few hundreds or thousands and several anchor nodes, the ranging procedures have the fatal disadvantage of processing delay and inefficient traffic bandwidth. On the other hand, the one-way ranging method is simple and fast, but susceptible to network synchronization. In this paper, we propose a method to modify asynchronous ranging equations to establish exact frequency or frequency offset, a method to estimate frequencies or frequency offsets, and a method to establish post-facto synchronization with anchor nodes. The synchronization for a node pair is adapted using instantaneous time information and corresponding difference of distances can be determined. We evaluate the performance of TWR, SDS-TWR and proposed ranging algorithms.

• 한국통신학회논문지
• /
• 제19권4호
• /
• pp.729-739
• /
• 1994

중앙집중 대역확산 패킷라디오 네트워크를 위한 다원접속 방식으로 임의할당-송신기 프로토콜을 새로이 제안한다. 제안하는 임의할당-송신기 프로토콜은 프리엠블 전송시만 경쟁모드로 동작하고, 데이터부분은 의사 직교특성을 갖는 송신기별 확산코드를 사용한다. 종래의 송신기별 프로토콜과 비교할때, 임의 할당-송신기 프로토콜은 프리엠블 전송을 위해 적은 수의 확산코드를 사용하는 경우 중앙수신기의 복잡도를 크게 줄일 수 있다. 이론적인 처리율 분석과 시뮬레이션 결과를 통해 단지 2~3개의 확산코드를 사용하여 프리엠블을 전송하는 경우에 성능이득의 대부분을 얻을 수 있었다.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제1권2호
• /
• pp.121-138
• /
• 2001

Due to large demand for high speed and great capacity for data transfer, WDM, which uses the wavelength division multiplexing technique, is known as alternative way to satisfy those demand for its flexible network operation and management, easy network expansion with existing networks, and enhancement of efficient data transfer rate. For these reasons, a new high capacity WDM optical communication network plan was established. Therefore, the quality of currently deployed optical cables with 81.6 km in length should be assessed to ensure if high capacity WDM system could be implemented on existing optical cables. Two important characteristic parameters, Transfer Loss and PMD (Polarization Mode Dispersion), were measured to evaluate quality of existing optical cable. Transfer Loss was measured at 0.244 dB per kilometer, which is lower than the design standard value at 0.275 dB/km. The measured PMD value gave at 0.030ps/km, and it, therefore, satisfies the value recommended by ITU-T (International Telecommunication Union-T) of 0.5ps/km. In addition, the transfer characteristic for existing 2.5 Gbps and 10 Gbps system were measured and evaluated, and the results showed that error-free transfer is very much feasible. Computer simulation for DWDM system, which is likely be a future backbone network in Korea, to assess the transfer characteristic using the same condition employed for 2.5 Gbps and 10 Gbps was carried out as well. The simulation verified that a stable network operation and reliable service could be provided.

• 경영과학
• /
• 제10권1호
• /
• pp.139-156
• /
• 1993

Various ATM switches have been proposed since Asynchronous Transfer Mode (ATM) was recognized as appropriate for implementing broadband integrated services digital network (BISDN). An ATM switching network may be evaluated on two sides : traffic performances (maximum throughput, delay, and packet loss probability, etc.) and structural features (complexity, i.e. the number of switch elements necessary to construct the same size switching network, maintenance, modularity, and fault tolerance, etc.). ATM switching networks proposed to date tend to show the contrary characteristics between structural features and traffic performance. The Knockout Switch, which is well known as one of ATM switches, shows a good traffic performance but it needs so many switch elements and buffers. In this paper, we propose an input and output buffered Knockout Switch for the purpose of reducing the number of switch elements and buffers of the existing Knockout Switch. We analyze the traffic performance and the structural features of the proposed switching architecture through a discrete time Markov chain and compare with those of the existing Knockout Switch. It was found that the proposed architecture could reduce more than 40 percent of switch elements and more than 30 percent of buffers under a given requirement of cell loss probability of the switch.

• Journal of Power Electronics
• /
• 제7권2호
• /
• pp.146-158
• /
• 2007

A multi-pulse GTO based voltage source converter (VSC) topology together with a fundamental frequency switching mode of gate control is a mature technology being widely used in static synchronous compensators (STATCOMs). The present practice in utility/industry is to employ a high number of pulses in the STATCOM, preferably a 48-pulse along with matching components of magnetics for dynamic reactive power compensation, voltage regulation, etc. in electrical networks. With an increase in the pulse order, need of power electronic devices and inter-facing magnetic apparatus increases multi-fold to achieve a desired operating performance. In this paper, a competitive topology with a fewer number of devices and reduced magnetics is evolved to develop an 18-pulse, 2-level $\pm$ 100MVAR STATCOM in which a GTO-VSC device is operated at fundamental frequency switching gate control. The inter-facing magnetics topology is conceptualized in two stages and with this harmonics distortion in the network is minimized to permissible IEEE-519 standard limits. This compensator is modeled, designed and simulated by a SimPowerSystems tool box in MATLAB platform and is tested for voltage regulation and power factor correction in power systems. The operating characteristics corresponding to steady state and dynamic operating conditions show an acceptable performance.

• ETRI Journal
• /
• 제43권3호
• /
• pp.436-446
• /
• 2021

In this paper, we propose an energy-efficient unmanned aerial vehicle (UAV) relaying network. In this network, the channels between UAVs and ground transceivers are model-free. A UAV acting as a flying relay explores better channels to assist in efficient data delivery between two ground nodes. The full-duplex relaying mode is applied for potential energy efficiency (EE) improvements. With the genetic algorithm, we manage to optimize the UAV trajectory for any arbitrary radio map scenario. Numerical results demonstrate that compared to other schemes (eg, fixed trajectory/speed policies), the proposed algorithm performs better in terms of EE. Additionally, the impact of self-interference on average EE is also investigated.

• ETRI Journal
• /
• 제45권1호
• /
• pp.163-170
• /
• 2023

We report an electroabsorption modulator (EAM)-integrated distributed Bragg reflector laser diode (DBR-LD) capable of supporting a high data rate and a wide wavelength tuning. The DBR-LD contains two tuning elements, plasma and heater tunings, both of which are implemented in the DBR section, which have blue-shift and red-shift in the Bragg wavelength through a current injection, respectively. The light created from the DBR-LD is intensity-modulated through the EAM voltage, which is integrated monolithically with the DBRLD using a butt-joint coupling method. The fabricated chip shows a threshold current of approximately 8 mA, tuning range of greater than 30 nm, and static extinction ratio of higher than 20 dB while maintaining a side mode suppression ratio of greater than 40 dB under a window of 1550 nm. To evaluate its modulation properties, the chip was bonded onto a mount including a radiofrequency line and a load resistor showing clear eye openings at data rates of 25 Gb/s nonreturn-to-zero and 50 Gb/s pulse amplitude modulation 4-level, respectively.

• Computers and Concrete
• /
• 제31권2호
• /
• pp.151-161
• /
• 2023

This paper presents an investigation of variables that cause spurious cracking in numerical modeling of concrete fracture. Spurious cracks appear due to the approximate nature of numerical modeling. They overestimate the dissipated energy, leading to divergent results with mesh refinement. This paper is limited to quasi-static loading regime, homogeneous models, cracking as the only nonlinear mode of deformation and cracking only due to tensile loading. Under these conditions, some variables that can be related to spurious cracking are: mesh alignment, ductility, crack band width, structure size, mesh refinement and load increment size. Case studies illustrate the effect of each variable and convergence analyses demonstrate that, after all, load-increment size is the most important variable. Theoretically, a sufficiently small load increment is able to eliminate or at least alleviate the detrimental influence of the other variables. Such load-increment size might be prohibitively small, rendering the simulation unfeasible. Hence, this paper proposes two alternatives. First, it is proposed an algorithm that automatically find such small load increment size automatically, which not necessarily avoid large computations. Then, it is proposed a double simulation technique, in which the crack is forced to propagate through the localization zone.