• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.11a
• /
• pp.358-361
• /
• 2004

This paper describes the development process of high stiffness body for ride and handling performance. High stiffness and light weight vehicle is a major target in the refinement of passenger cars to meet customers' contradictable requirements between ride and handling performance and fuel economy. This paper describes the analysis approach process for high stiffness body through the data level of body stiffness. According to the frequency band, we can suggest the design guideline about Is cornering static stiffness, torsional and lateral stiffness, body attachment stiffness. The ride and handling characteristic of a vehicle is significantly affected by vibration transferred to the body through the chassis mounting points from front and rear suspension. It is known that body attachment stiffness is an important factor of ride and handling performance improvement. And high stiffness helps to improve the flexibility of bushing rate tuning between Handling and road noise. It makes it possible to design the good handling performance vehicle at initial design stage and save vehicles to be used in tests by using mother car at initial design stage. These improvements can lead to shortening the time needed to develop better vehicles.

• Journal of KIISE:Databases
• /
• v.33 no.1
• /
• pp.117-128
• /
• 2006

In this paper, we propose a streaming method for XML data that supports energy-efficient processing of queries over the stream in mobile clients. We propose new stream organizations for XML data which have different kinds of addresses to related data in a stream. We describe event-driven stream generation algorithms for the proposed stream structures and provide search algorithms for simple XML path queries which leverage the access mechanisms incorporated in the stream. Experimental results show that our approaches can effectively improve the tuning time performance of user queries in a wireless broadcasting environment.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2012.05a
• /
• pp.349-350
• /
• 2012

Porous silicon (PS) was prepared by electrochemical anodization. Ultra-thin zinc oxide (ZnO) capping layers were deposited on the PS by plasma-assisted molecular beam epitaxy (PA-MBE). The effects of the ZnO capping layers on the properties of the as-prepared PS were investigated using scanning electron microscopy (SEM) and photoluminescence (PL). The as-prepared PS has circular pores over the entire surface. Its structure is similar to a sponge where the quantum confinement effect (QCE) plays a fundamental role. It was found that the dominant red emission of the porous silicon was tuned to white light emission by simple deposition of the ultra-thin ZnO capping layers. Specifically, the intensity of white light emission was observed to be enhanced by increasing the growth time from 1 to 3 min.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2008.03a
• /
• pp.871-876
• /
• 2008

The blade vibration problem of bladed disk is the most critical subject to consider since it directly affects the stable performance of the engine as well as life of the engine. Especially, due to complicated vibration pattern of the bladed disk, more effort was required for vibration analysis and test. The research of measuring the vibration of the bladed disk, using NSMS(Non-intrusive stress measurement) instead of Aeromechanics testing method requiring slip ring or telemetry system with strain gauge, was successful. These testing can report the actual stresses seen on the blades; detect synchronous resonances that are the source of high cycle fatigue(HCF) in blades; measure individual blade mis-tuning and coupled resonances in bladed disks. In order to minimize the error being created due to heat expansion, the tip timing sensor is installed parallel to the blade trailing edge, yielding optimal result. Also, when working on finite element analysis, the whole bladed disk has gone through three-dimensional analysis, evaluating the family mode. The result of the analysis matched well with the test result.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1795-1811
• /
• 2019

In order to achieve rapid and accurate detection of vehicle objects in complex traffic conditions, we propose a novel vehicle detection method. Firstly, more contextual and small-object vehicle information can be obtained by our Joint Feature Network (JFN). Secondly, our Evolved Region Proposal Network (EPRN) generates initial anchor boxes by adding an improved version of the region proposal network in this network, and at the same time filters out a large number of false vehicle boxes by soft-Non Maximum Suppression (NMS). Then, our Mask Network (MaskN) generates an example that includes the vehicle occlusion, the generator and discriminator can learn from each other in order to further improve the vehicle object detection capability. Finally, these candidate vehicle detection boxes are optimized to obtain the final vehicle detection boxes by the Fine-Tuning Network(FTN). Through the evaluation experiment on the DETRAC benchmark dataset, we find that in terms of mAP, our method exceeds Faster-RCNN by 11.15%, YOLO by 11.88%, and EB by 1.64%. Besides, our algorithm also has achieved top2 comaring with MS-CNN, YOLO-v3, RefineNet, RetinaNet, Faster-rcnn, DSSD and YOLO-v2 of vehicle category in KITTI dataset.

• Current Optics and Photonics
• /
• v.3 no.1
• /
• pp.27-32
• /
• 2019

We present an autofocus tracking system implemented by the digital refocusing of digital holographic microscopy (DHM) and the tunability of an electrically tunable lens (ETL). Once the defocusing distance of an image is calculated with the DHM, then the focal plane of the imaging system is optically tuned so that it always gives a well-focused image regardless of the object location. The accuracy of the focus is evaluated by calculating the contrast of refocused images. The DHM is performed in an off-axis holographic configuration, and the ETL performs the focal plane tuning. With this proposed system, we can easily track down the object drifting along the depth direction without using any physical scanning. In addition, the proposed system can simultaneously obtain the digital hologram and the optical image by using the RGB channels of a color camera. In our experiment, the digital hologram is obtained by using the red channel and the optical image is obtained by the blue channel of the same camera at the same time. This technique is expected to find a good application in the long-term imaging of various floating cells.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.475-482
• /
• 2018

In this paper, we present a study on data augmentation methods for DNN (Deep Neural Network)-based audio tagging. In this system, an audio signal is converted into a mel-spectrogram and used as an input to the DNN for audio tagging. To cope with the problem associated with a small number of training data, we augment the training samples using time stretching, pitch shifting, dynamic range compression, and block mixing. In this paper, we derive optimal parameters and combinations for the augmentation methods through audio tagging simulations.

• Proceedings of the KIPE Conference
• /
• 2019.07a
• /
• pp.71-73
• /
• 2019

The output current of the Grid Connected Inverter (GCI) can be polluted with harmonics mainly due to i) dead time in switches, ii) non-linearity of switches, iii) grid harmonics, and iv) DC link fluctuation. Therefore, it is essential to design the robust Harmonic Compensation (HC) technique for the improvement of output current quality and fulfill the IEEE 1547 Total harmonics Distortion (THD) limit i.e. <5%. The conventional harmonic techniques often are complex in implementation due to their i) additional hardware needs, ii) complex structure, iii) difficulty in tuning of parameters, iv) current controller compatibility issues, and v) higher computational burden. In this paper, to eliminate the harmonics from the GCI output current, a novel Digital Lock-In Amplifier (DLA) based harmonic detection is proposed. The advantage of DLA is that it extracts the harmonic information accurately, which is further compensated by means of PI controller in feed forward manner. Moreover, the proposed HC method does not require additional hardware and it works with any current controller reference frame. To show the effectiveness of the proposed HC method a 5kW GCI prototype built in laboratory. The output current THD is achieved less than 5% even with 10% load, which is verified by simulation and experiment.

• KEPCO Journal on Electric Power and Energy
• /
• v.4 no.2
• /
• pp.89-99
• /
• 2018

Downtime and malfunction of industrial rotor machines represents a crucial cost burden and productivity loss. Fault diagnosis of this equipment has recently been carried out to detect their fault(s) and cause(s) by using fault classification methods. However, these methods are of limited use in detecting rotor faults because of their hypersensitivity to unexpected and different equipment conditions individually. These limitations tend to affect the accuracy of fault classification since fault-related features calculated from vibration signal are moved to other regions or changed. To improve the limited diagnosis accuracy of existing methods, we propose a new approach for fault diagnosis of rotor machines based on the model generated by supervised learning. Our work is based on feature residual values from vibration signals as fault indices. Our diagnostic model is a robust and flexible process that, once learned from historical data only one time, allows it to apply to different target systems without optimization of algorithms. The performance of the proposed method was evaluated by comparing its results with conventional methods for fault diagnosis of rotor machines. The experimental results show that the proposed method can be used to achieve better fault diagnosis, even when applied to systems with different normal-state signals, scales, and structures, without tuning or the use of a complementary algorithm. The effectiveness of the method was assessed by simulation using various rotor machine models.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.6-10
• /
• 2021

Recognition of characters printed or marked on the PCB components from images captured using cameras is an important task in PCB components inspection systems. Previous optical character recognition (OCR) of PCB components typically consists of two stages: character segmentation and classification of each segmented character. However, character segmentation often fails due to corrupted characters, low image contrast, etc. Thus, OCR without character segmentation is desirable and increasingly used via deep neural networks. Typical implementation based on deep neural nets without character segmentation includes convolutional neural network followed by recurrent neural network (RNN). However, one disadvantage of this approach is slow execution due to RNN layers. LPRNet is a segmentation-free character recognition network with excellent accuracy proved in license plate recognition. LPRNet uses a wide convolution instead of RNN, thus enabling fast inference. In this paper, LPRNet was adapted for recognizing characters printed on PCB components with fast execution and high accuracy. Initial training with synthetic images followed by fine-tuning on real text images yielded accurate recognition. This net can be further optimized on Intel CPU using OpenVINO tool kit. The optimized version of the network can be run in real-time faster than even GPU.