• Wind and Structures
• /
• v.34 no.6
• /
• pp.511-523
• /
• 2022

Detecting the icing on wind turbine blades built-in cold regions with conventional methods is always a very laborious, expensive and very difficult task. Regarding this issue, the use of smart systems has recently come to the agenda. It is quite possible to eliminate this issue by using the deep learning method, which is one of these methods. In this study, an application has been implemented that can detect icing on wind turbine blades images with visualization techniques based on deep learning using images. Pre-trained models of Resnet-50, VGG-16, VGG-19 and Inception-V3, which are well-known deep learning approaches, are used to classify objects automatically. Grad-CAM, Grad-CAM++, and Score-CAM visualization techniques were considered depending on the deep learning methods used to predict the location of icing regions on the wind turbine blades accurately. It was clearly shown that the best visualization technique for localization is Score-CAM. Finally, visualization performance analyses in various cases which are close-up and remote photos of a wind turbine, density of icing and light were carried out using Score-CAM for Resnet-50. As a result, it is understood that these methods can detect icing occurring on the wind turbine with acceptable high accuracy.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.4
• /
• pp.290-298
• /
• 2013

A walking navigation system (usually known as a locomotion interface) is an interactive platform which gives simulated walking sensation to users using sensed bipedal motion signals. This enables us to perform navigation tasks using only bipedal movement. Especially, it is useful for the certain VR task which emphasizes on physical human movement, or accompanies understanding of the size and complexity of building structures. In this work, we described system components of VR walking system and investigated several types of walking platform by literature survey. We adopted a MS Kinect depth sensor for the motion recognition and a treadmill which includes directional turning mechanism for the walking platform. Through the integration of these components with a VR navigation scenario, we developed a simple VR walking navigation system. Finally several technical issues were found during development process, and further research directions were suggested for the system improvement.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.2
• /
• pp.154-161
• /
• 1998

Prototypes of a design are always needed for the purpose of visualization and evaluation in the aspect of manufacturability functionality, and aesthetic appearance. Since the prototyping process requires a significant amount of cost and time, various rapid prototyping processes are recently being introduced in the process. However, it is usually necessary for a part built up by a rapid prototyping system to be refined by a post-processing process, in which the stair steps on the surfaces, the support structures (if they exist), and the unprocessed material are eliminated. This post-processing is usually done manually and is a time-consuming task. Especially, eliminating the trapped volumes, the volume of the unprocessed material entrapped by the solidified portion, is sometimes impossible in some processes. This study provides a designer with a tool to detect the existence and to calculate the quantity of the trapped volumes at the given build-up direction, so that the proper build-up direction is chosen or the part is built by pieces to avoid the problems caused by the trapped volumes in advance. Since the proposed algorithm can efficiently calculate the amount of the trapped volumes at any build-up direction, it has the potential of such application as optimizing the build-up direction to minimize the trapped volumes.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.06a
• /
• pp.136-147
• /
• 2002

The purpose of this study is to understand what we need to know about Microsoft's powerful web browser control so that we will be able to code and create a special-purpose browser. For this purpose, first of all, some literature survey was undertaken to study how the web browser looks like and is operated. One case design and coding was devoted to using visual basic to create our own web browser. The primary task of this job was accomplished with the form making up of a bunch of image controls, a status bar on the button, a picture box to contain the image buttons, the combo box, the command button and the menus and the statements shown in the above exhibits and tables respectively. After putting it all together, the form was implemented to see if it would work properly. The program was also enhanced to have a special purpose functions permitting filtering functions. The result was more or less successful, however has some limitations such as the simplicity of the programming structures and functions. However, this study is useful for understanding general outline of special-purpose web browser development.

• Wind and Structures
• /
• v.32 no.3
• /
• pp.193-204
• /
• 2021

The estimation of the extreme wind load (effect) under a mean recurrence interval (MRI) is an important task in the wind-resistant design for the structure. It can be predicted by either first-order method or full-order method, depending on the accuracy and complexity requirement. Although the first-order method with the consideration of wind directionality has been proposed, less work has been done on the full-order method, especially with the wind directionality. In this study, the full-order method considering the wind directionality is proposed based on multivariate joint probability distribution. Meanwhile, considering two wind directions, the difference of the corresponding results based on the first-order method and full-order method is analyzed. Finally, based on the measured wind speed data, the discrepancy between these two methods is investigated. Results show that the difference between two approaches is not obvious under larger MRIs while the underestimation caused by the first-order method can be larger than 15% under smaller MRIs. Overall, the first-order method is sufficient to estimate the extreme wind load (effect).

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2004.04a
• /
• pp.423-430
• /
• 2004

A new active neuro-control strategy for seismic response reduction using modal states is proposed. In order to apply the neuro-control strategy to the given structural system it is needed to select state variables used as inputs into the neural network. If the degrees of freedom of the analytical model is large, there are so many possible combinations of the state variables. And selecting state variables is very complicated and troublesome task for the designer. In order to avoid this problem, the proposed control system adopts modal states as inputs. Since the modal states contain the information of the whole structural system's behavior, it is proper to use modal states as inputs of the neuro-controller. The simulation results show that the proposed the proposed active neuro-control strategy is quite effective to reduce seismic responses. In addition, the consuming time for training proposed neuro-controller is quite shorter than that for the conventional neuro- controller. The results of this investigation, therefore, indicate that the proposed active neuro-control strategy using modal states as the inputs could be effectively used for control seismically excited structures.

• Carbon letters
• /
• v.11 no.3
• /
• pp.184-191
• /
• 2010

Carbon nanotubes are unique tubular structures of nanometer diameter and large length/diameter ratio. The nanotubes may consist of one up to tens and hundreds of concentric shells of carbons with adjacent shells separation of ~0.34 nm. Multiwalled carbon nanotubes were synthesized by arc-discharge technique. MWCNTs were formed at the cathode deposit along with other carbonaceous materials like amorphous carbon, graphite etc. However, to get the best advantage of carbon nanotubes in various advanced applications, these undesired carbonaceous materials to be removed which is a challenging task. In the present study, various techniques were tried out for purifying MWCNTs such as physical filtration, chemical treatment and thermal annealing. SEM, FTIR, TGA and BET techniques were used to characterize the CNTs at various stages. Results shows that suitable chemical treatment followed by thermal annealing under controlled flow of oxygen gives the better route for purification of carbon nanotubes.

• Earthquakes and Structures
• /
• v.3 no.6
• /
• pp.839-852
• /
• 2012

Earthquake records are often analyzed in various earthquake engineering problems, making time-frequency analysis for such records of primary concern. The best tool for such analysis appears to be based on wavelet functions; selection of which is not an easy task and is commonly carried through trial and error process. Furthermore, often a particular wavelet is adopted for analysis of various earthquakes irrespective of record's prime characteristics, e.g. wave's magnitude. A wavelet constructed based on records' characteristics may yield a more accurate solution and more efficient solution procedure in time-frequency analysis. In this study, a low-pass reconstruction filter is obtained for each earthquake record based on multi-resolution decomposition technique; the filter is then assigned to be the normalized version of the last approximation component with respect to its magnitude. The scaling and wavelet functions are computed using two-scale relations. The calculated wavelets are highly efficient in decomposing the original records as compared to other commonly used wavelets such as Daubechies2 wavelet. The method is further advantageous since it enables one to decompose the original record in such a way that a clear time-frequency resolution is obtained.

• Smart Structures and Systems
• /
• v.17 no.4
• /
• pp.631-646
• /
• 2016

Although the deployment of wireless sensors for structural sensing and monitoring is becoming popular, supplying power to these sensors remains as a daunting task. To address this issue, there have been large volume of ongoing energy harvesting studies that aimed to find a way to scavenge energy from surrounding ambient energy sources such as vibration, light and heat. In this study, a magnetic resonance based wireless power transfer (MR-WPT) system is proposed so that sensors inside a concrete structure can be wirelessly powered by an external power source. MR-WPT system offers need-based active power transfer using an external power source, and allows wireless power transfer through 300-mm thick reinforced concrete with 21.34% and 17.29% transfer efficiency at distances of 450 mm and 500 mm, respectively. Because enough power to operate a typical wireless sensor can be instantaneously transferred using the proposed MR-WPT system, no additional energy storage devices such as rechargeable batteries or supercapacitors are required inside the wireless sensor, extending the expected life-span of the sensor.