• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.383-392
• /
• 2023

Post-earthquake building condition assessment is crucial for subsequent rescue and remediation and can be automated by emerging computer vision and deep learning technologies. This study is based on an endeavour for the 2nd International Competition of Structural Health Monitoring (IC-SHM 2021). The task package includes five image segmentation objectives - defects (crack/spall/rebar exposure), structural component, and damage state. The structural component and damage state tasks are identified as the priority that can form actionable decisions. A multi-task Convolutional Neural Network (CNN) is proposed to conduct the two major tasks simultaneously. The rest 3 sub-tasks (spall/crack/rebar exposure) were incorporated as auxiliary tasks. By synchronously learning defect information (spall/crack/rebar exposure), the multi-task CNN model outperforms the counterpart single-task models in recognizing structural components and estimating damage states. Particularly, the pixel-level damage state estimation witnesses a mIoU (mean intersection over union) improvement from 0.5855 to 0.6374. For the defect detection tasks, rebar exposure is omitted due to the extremely biased sample distribution. The segmentations of crack and spall are automated by single-task U-Net but with extra efforts to resample the provided data. The segmentation of small objects (spall and crack) benefits from the resampling method, with a substantial IoU increment of nearly 10%.

• International Journal of Automotive Technology
• /
• 제7권2호
• /
• pp.145-154
• /
• 2006

In the companion paper, the composition and structure of the MATDYMO (Multi-Agent for Traffic Simulation with Vehicle Dynamic Model) were proposed. MATDYMO consists of the road management system, the vehicle motion control system, the driver management system, and the integration control system. Among these systems, the road management system and the integration control system were discussed In the companion paper. In this paper, the vehicle motion control system and the driver management system are discussed. The driver management system constructs the driver agent capable of having different driving styles ranging from slow and careful driving to fast and aggressive driving through the yielding index and passing index. According to these indices, the agents pass or yield their lane for other vehicles; the driver management system constructs the vehicle agents capable of representing the physical vehicle itself. A vehicle agent shows its behavior according to its dynamic characteristics. The vehicle agent contains the nonlinear subcomponents of engine, torque converter, automatic transmission, and wheels. The simulation is conducted for an interrupted flow model and its results are verified by comparison with the results from a commercial software, TRANSYT-7F. The interrupted flow model simulation is implemented for three cases. The first case analyzes the agents' behaviors in the interrupted flow model and it confirms that the agent's behavior could characterize the diversity of human behavior and vehicle well through every rule and communication frameworks. The second case analyzes the traffic signals changed at different intervals and as the acceleration rate changed. The third case analyzes the effects of the traffic signals and traffic volume. The results of these analyses showed that the change of the traffic state was closely related with the vehicle acceleration rate, traffic volume, and the traffic signal interval between intersections. These simulations confirmed that MATDYMO can represent the real traffic condition of the interrupted flow model. At the current stage of development, MATDYMO shows great promise and has significant implications on future traffic state forecasting research.

• Journal of Power Electronics
• /
• 제13권4호
• /
• pp.569-583
• /
• 2013

In this study, the simultaneous use of a multi-level converter (MLC) as a DC-motor drive and as an active battery cell balancer is investigated. MLCs allow each battery cell in a battery pack to be independently switched on and off, thereby enabling the potential non-uniform use of battery cells. By exploiting this property and the brake regeneration phases in the drive cycle, MLCs can balance both the state of charge (SoC) and temperature differences between cells, which are two known causes of battery wear, even without reciprocating the coolant flow inside the pack. The optimal control policy (OP) that considers both battery pack temperature and SoC dynamics is studied in detail based on the assumption that information on the state of each cell, the schedule of reciprocating air flow and the future driving profile are perfectly known. Results show that OP provides significant reductions in temperature and in SoC deviations compared with the uniform use of all cells even with uni-directional coolant flow. Thus, reciprocating coolant flow is a redundant function for a MLC-based cell balancer. A specific contribution of this paper is the derivation of a state-space electro-thermal model of a battery submodule for both uni-directional and reciprocating coolant flows under the switching action of MLC, resulting in OP being derived by the solution of a convex optimization problem.

• Nuclear Engineering and Technology
• /
• 제51권5호
• /
• pp.1218-1230
• /
• 2019

A sub-channel solver, named ${\underline{S}}teady$ and ${\underline{T}}ransient$ ${\underline{A}}nalyzer$ for ${\underline{R}}eactor$ ${\underline{T}}hermal$ hydraulics (START), has been developed using the homogenous model for two-phase conditions of light water reactors. The code is developed as a fast and accurate TH-solver for coupled and multi-physics calculations. START has been validated against the NUPEC PWR Sub-channel and Bundle Test (PSBT) database. Tests like single-channel quality and void-fraction for steady state, outlet fluid temperature for steady state, rod-bundle quality and void-fraction for both steady state and transient conditions have been analyzed and compared with experimental values. Results reveal a good accuracy of solution for both steady state and transient scenarios. Axially different values for turbulent mixing coefficient are used based on different grid-spacer types. This provides better results as compared to using a single value of turbulent mixing coefficient. Code-to-code evaluation of PSBT results by the START code compares well with other industrial codes. The START code has been parallelized with the OpenMP algorithm and its numerical performance is evaluated with a large whole PWR core. Scaling study of START shows a good parallel performance.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 춘계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.106-109
• /
• 2008

Piecewise linear diode models are widely used for large-signal circuit analyses, especially power electronic circuit simulations. When using a piecewise linear diode model for simulation, a switching method to select a proper one among linear models is needed. The conventional switching method keeps the previous ON, OFF state information, and applies different switching conditions according to the state. However, this method has difficulties especially in extending to multi-piecewise linear models. This paper presents a switching method which appropriately divides the v-i plane into regions and select a linear model according to the region where the operating point(the voltage and the current of the diode) belongs. This switching method is easily extended to multi-Piecewise linear models. An example using the tableau analysis and the backward Euler integration is presented for verification.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 추계학술대회 논문집 전력기술부문
• /
• pp.19-21
• /
• 2008

Renewable energy resources such as wind, wave, solar, micro hydro, tidal and biomass etc. are becoming importance stage by stage because of considering effect of the environment. Solar energy is one of the most successful sources of renewable energy for the production of electrical energy following wind energy. And, the solar/photovoltaic cell generators can not make two-state model as conventional generators, but should be modeled as multi-state model due to solar radiation random variation. The method of obtaining reliability evaluation index of solar cell generators is different from the conventional generators. This paper presents a basic study on reliability evaluation of power system considering solar cell generators with multi-states.

• 한국철도학회논문집
• /
• 제12권2호
• /
• pp.285-290
• /
• 2009

논문에서는 연속HMM(Continuos Hidden Markov Model)을 이용하여 실시간으로 발매기 명령어(314개 역명)를 인식 할 수 있도록 음성인식 시스템을 구현하였다. 특징 벡터로 39 MFCC를 사용하였으며, 인식률 향상을 위하여 895개의 tied-state 트라이폰 음소 모델을 구성하였다. 시스템 성능 평가 결과 다중 화자 종속 인식률은 99.24%, 다중화자 독립 인식률은 98.02%의 인식률을 나타내었으며, 실제 노이즈가 있는 환경에서 다중 화자 독립 실험의 경우 93.91%의 인식률을 나타내었다.

• ETRI Journal
• /
• 제29권5호
• /
• pp.688-690
• /
• 2007

This letter presents a filter for a car navigation system integrating a low-grade inertial measurements unit (IMU) and a global positioning system receiver. The filter is designed according to the state variables to be estimated and the usable measurements. The usable measurements change from case to case, and the estimative state variables also change due to the measurements; therefore, multiple models must be used for real environmental maneuvers. In this letter, four models for land navigation are chosen and switched by rearranging the system matrix and resetting the error covariance matrices.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 A
• /
• pp.209-211
• /
• 2000

This paper presents a Multi-Agent system for setting and coordination of overcurrent protective devices in the distribution system. The simulation model shows its feasibility in which KQML is used for communication among agents. The proposed system could make it possible to keep the protection system in the optimal state all the time.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2005년도 학술발표회(1)
• /
• pp.538-539
• /
• 2005