• KIEAE Journal
• /
• 제7권1호
• /
• pp.23-31
• /
• 2007

This study aims to evaluate the illuminance performance of sloped light guide with automated venetian blind systems. For the purpose, a mock-up model was constructed as a prototype of Korean office building with $12.0m{\times}7.3m{\times}3.7m$ ($w{\times}d{\times}h$) and south facing side-window mounted between the clerestory window($2.0m^2$) and the view window($5.6m^2$). The light guide of 1.28m deepth and $29^{\circ}$ tilted angle, is covered with 0.6mm galvanized steel sheet and 97% reflective film. To protect the room from low solar angle, a blind systems, 0.15m deepth and $30^{\circ}$ automated slat angle was installed. To assess illuminance performance, the totally 37 measuring points for illuminance were monitored. For the detailed analysis, photometric sensors were installed at work-plane (8 points), wall (7 points), ceiling (3points), and exterior horizontal illuminance (1 point) respectively. The performance was measured under clear sky and is monitored by Agilent data logger, photometric sensor Li-cor and the Radiant Imaging ProMetric 1400. Comparisons of light factor and uniformity are discussed.

• KIEAE Journal
• /
• 제7권1호
• /
• pp.65-72
• /
• 2007

As ecological design elements, daylighting can be applied to provide adequate illumination on visual tasks to create an attractive visual environment and to save electrical energy. Daylighting control systems reject direct sunlight and penetrate it onto the ceiling or to deep into the room. This study aims to evaluate the luminance environment of sloped light guide with automated venetian blind systems according to sun angle changes. For evaluation, a mock-up model was used and the south facing side-window mounted between the clerestory window and the view window. To assess luminance performance, 3 view points of luminance were monitored. As results, the conventional and lightshelves show ideal luminance ratio between workplane and surroundings(3:1) and workplane and darkness area(2:1) due to total ratio of surroundings and darkness area has lower ratio than workplane. Compared to the lightshelves window, conventional window shows unrelieved effect in between the workplane and brightness area(1:5). It means that there has low deviation according to the required standards. Also, compared to the ratio between the brightness area and darkness area(2~6:1) conventional window with high deviation(10~20:1) provide discomfort glare due to the excessively strong contrast, while lightshelves window shows a required luminance ratio that provide a three-dimensional effect to occupants. Therefore, luminance distribution indicate that application of a lightshelves and blinds not only has a significantly positive effect but also offers higher luminance quality in a daylit room

• 조명전기설비학회논문지
• /
• 제23권12호
• /
• pp.1-9
• /
• 2009

최근 실내에 유입된 주광을 활용하여 인공조명의 에너지를 절감하는 광센서 조광제어시스템에 관한 연구는 유용한 주광의 유입을 위한 자동 차양장치와의 통합을 중심으로 이루어지고 있다. 기존의 연구는 자동 롤러쉐이딩시스템과 베니션블라이드에 관한 연구로, 이 두 시스템은 주광의 입사를 수직적으로 계산한 일영각을 기준으로 차양장치의 높이 및 각도를 계산하여 제어하고 있다. 본 연구는 광센서 조광제어시스템과 연동을 위한 새로운 차양장치로 주광의 유입을 수직 수평으로 제어하는 복층유리 격자루버시스템을 제안하고, 창면의 위치에 따른 루버의 간격을 산출하였으며, 컴퓨터 시뮬레이션을 통하여 격자루버시스템의 주광유입특성을 분석하고 평가하였다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.