• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.10
• /
• pp.1281-1292
• /
• 2001

Experiments were done for the comparison of performance and flow characteristics between a two -stage axial flow fan and a counter-rotating axial flow fan. Each stage of the two -stage axial flow fan used fur the present study has an eight bladed rotor and thirteen slater blades. The front and the rear rotor of the counter - rotating axial flow fan have eight blades each and are driven by coaxial counter ro latins shafts through a gearbox located between the rear rotor and the electric motor. Both of the two axial fan configurations have identical rotor blades and the same operating condition fur the one -to-one comparison of the two. Performance curves of the two configurations were obtained and compared by varying the blade pitch angles and axial gaps between the blade rows. The fan characteristic curves were obtained following the Korean Standard Testing Methods for Turbo Fans and Blowers (KS B 6311). The fa n flow characteristics were measured using a five-hole probe by a non-nulling method. The velocity profiles between the hub and tip of the fans were measured and analyzed at the particular operating condition s of peak efficiency, minimum and maximum pressure coefficients. The peak efficiency of the counter-rotating axial fan was improved about 2% respectively, compared with the two stage axial fan. At the minimum pressure coefficient point of the two stage axial fan, the fan inlet flow patterns show that axial velocity highly decreased in the vicinity of the blade tip region. Also, the reverse flow took place at the blade tip.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.1
• /
• pp.42-48
• /
• 2006

In this paper a joint detection method for other cell interference cancellation is proposed in the next generation hybrid TD-CDMA mobile communication systems. A joint detection technique, a most characteristic feature of hybrid TD-CDMA mobile communication systems. retrieves users' data in the same time slot simultaneously with the elimination of multiple user interference. Previously a two stage joint detection method was proposed to cancel other cell interference as well as multiple user interference in the target cell. However the previous scheme does not have concrete ways to recognize other cell users who give major interference to the target cell. Thus all users in neighbor other cells has to be jointly detected and it causes huge complexity of the two stage joint detection. In this paper a method is proposed to perform two stage joint detection according to users' interference with the target cell. Performances of the proposed scheme are investigated through simulations and compared to the previous method the proposed method has no performance degradation and also lower the complexity of two stage joint detection significantly.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.7C
• /
• pp.719-725
• /
• 2003

The content-based image retrieval system extracts features of color, shape and texture from raw images, and builds the database with those features in the indexing process. The search in the whole retrieval system is defined as a process which finds images that have large similarity to query image using the feature database. This paper proposes a new two-stage search method in the content-based image retrieval system. The method is that the features are condensed and stored by the property of Cauchy-Schwartz inequality in order to reduce the similarity computation time which takes a mostly response time from entering a query to getting retrieval results. By the extensive computer simulations, we have observed that the proposed two-stage search method successfully reduces the similarity computation time while maintaining the same retrieval relevance as the conventional exhaustive search method. We also have observed that the method is more effective as the number of images and dimensions of the feature space increase.

• Journal of the Korea Convergence Society
• /
• v.12 no.7
• /
• pp.169-179
• /
• 2021

This study shows that the two-stage k-means clustering method can improve prediction performance by predicting the stock price, To this end, this study introduces the two-stage k-means clustering algorithm and tests the prediction performance through comparison with various machine learning techniques. It selects the cluster close to the prediction target obtained from the k-means clustering, and reapplies the k-means clustering method to the cluster to search for a cluster closer to the actual value. As a result, the predicted value of this method is shown to be closer to the actual stock price than the predicted values of other machine learning techniques. Furthermore, it shows a relatively stable predicted value despite the use of a relatively small cluster. Accordingly, this method can simultaneously improve the accuracy and stability of prediction, and it can be considered as the new clustering method useful for small data. In the future, developing the two-stage k-means clustering is required for the large-scale data application.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.1 s.307
• /
• pp.53-66
• /
• 2006

This paper proposes the combined image retrieval system that gives the same relevance as exhaustive search method while its performance can be considerably improved. This system is combined with two different retrieval methods and each gives the same results that full exhaustive search method does. Both of them are two-stage method. One uses condensation of feature vectors, and the other uses binary-tree clustering. These two methods extract the candidate images that always include correct answers at the first stage, and then filter out the incorrect images at the second stage. Inasmuch as these methods use equal algorithm, they can get the same result as full exhaustive search. The first method condenses the dimension of feature vectors, and it uses these condensed feature vectors to compute similarity of query and images in database. It can be found that there is an optimal condensation ratio which minimizes the overall retrieval time. The optimal ratio is applied to first stage of this method. Binary-tree clustering method, searching with recursive 2-means clustering, classifies each cluster dynamically with the same radius. For preserving relevance, its range of query has to be compensated at first stage. After candidate clusters were selected, final results are retrieved by computing similarities again at second stage. The proposed method is combined with above two methods. Because they are not dependent on each other, combined retrieval system can make a remarkable progress in performance.

• Journal of Global Scholars of Marketing Science
• /
• v.13
• /
• pp.135-158
• /
• 2004

On-line consumers have some visit behavior characteristics when they visit internet-shopping mall between visit-stage and purchase-stage. Therefore, information of on-line consumers have influenced on internet-shopping mall's profitabilities at site manager's perspectives. For examples, Are any on-line consumers continuous visiting under any situations? Or are any on-line consumers purchased on any specific internet-shopping mall? Expecially in this paper, researcher tried to understand visit behavioral characteristics of on-line consumers using two-stage mixture regression analysis. Throughout this process, it could be proposed method, which could be reinforced competitiveness of internet-shopping mall by segmental decision-making method. Additionally, it is expected that visit behavioral characteristics' information could be supplied strategic implications between visit-stage and purchase-stage Throughout empirical test it could be proved two-stage decision-making process, which decision-making process of on-line consumers would be processed visit-stage and purchase-stage. In this study, researcher proposed suitable response strategy after understanding visiting behavioral characteristics of on-line consumers. This paper has some academical contributions, which visit behavioral characteristics of on-line consumers could be grasped the meaning by site stickiness and navigation pattern.

• KSBB Journal
• /
• v.15 no.5
• /
• pp.489-496
• /
• 2000

Several culture systems including batch, two-stage CSTR, semi-fed batch, and two-stage cyclic fed-batch were investigated for the efficient production of the Fab fraction of PDC-E2 specific human monoclonal antibody using high cell density recombinant E. coli. A two-phase batch system and a two-stage continuous system were examined to overcome plasmid instability problems, by separating the growth and the production stages. The cell density and productivity of the two-stage continuous culture was better than that of the two-phase batch fermentation. In the two-stage continuous culture system with DO-stat, the cell growth and the productivity were superior to those of the system without the DO control. Also, almost total plasmid stability was maintained in the two-stage continuous culture system. Modified M9 medium was selected as an optimum feeding medium for the fed-batch process, and the optimum C/N ratio determined to be 2:3. The optimum feeding rate was $0.6g/\ell/hr$ for a constant feeding strategy in semi-fed batch system. When the feeding medium was fed by pulsing, it was observed that more frequent pulsing resulted in improved cell growth. The linear feeding method was the most efficient of the various feeding methods tested. Finally, high cell density culture using a two-stage cyclic fed batch system with pH-stat was tried because the linear feeding method showed limitations in terms of obtaining high cell densities, and a cell density of $54 g/\ell$ was achieved. It was concluded that the two-stage cyclic fed batch system was the most efficient system for high cell density culture of the systems tested. However, productivity improvements were lower than expected due to the extremely high accumulations of acetate, although the low levels of residual glucose were maintained.

• Computers and Concrete
• /
• v.12 no.2
• /
• pp.169-186
• /
• 2013

The aim of this paper is to determine the effect of soil-structure interaction and time dependent material properties on behavior of concrete box-girder highway bridges. Two different finite element analyses, one stage and construction stage, have been carried out on Komurhan Bridge between Elazi$\breve{g}$ and Malatya province of Turkey, over Fırat River. The one stage analysis assume that structure was built in a second and material properties of structure not change under different loads and site conditions during time. However, construction stage analysis considers that construction time and time dependent material properties. The main and side spans of bridge are 135 m and 76 m, respectively. The bridge had been constructed in 3 years between 1983 and 1986 by balanced cantilever construction method. The parameters of soil-structure interaction (SSI), time dependent material properties and construction method are taken into consideration in the construction stage analysis while SSI is single parameter taking into consideration in the one stage analysis. The 3D finite element model of bridge is created the commercial program of SAP2000. Time dependent material properties are elasticity modulus, creep and shrinkage for concrete and relaxation for steel. Soft, medium, and firm soils are selected for evaluating SSI in both analyses. The results of two different finite element analyses are compared with each other. It is seen that both construction stage and SSI have a remarkable effect on the structural behavior of the bridge.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.7B
• /
• pp.892-896
• /
• 2011

This paper proposes a two-stage synchronization scheme using a serial search non-coherent correlator appropriate for the IEEE 802.15.4a system. The proposed method improved the synchronization performance by using multi-pulse signals unlike the conventional method using single-pulse signals. It also compensated for the degradation of performance at low SNR resulting from the use of fixed threshold by applying the adaptive threshold technique. The proposed method showed a detection probability that is higher by approximately 0.2-0.3 compared with the conventional method in the IEEE 802.15.4a channel model.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.29-39
• /
• 2022

Crack detection plays an important role in the maintenance and protection of steel box girder of bridges. However, since the cracks only occupy an extremely small region of the high-resolution images captured from actual conditions, the existing methods cannot deal with this kind of image effectively. To solve this problem, this paper proposed a novel three-stage method based on deep learning technology and morphology operations. The training set and test set used in this paper are composed of 360 images (4928 × 3264 pixels) in steel girder box. The first stage of the proposed model converted high-resolution images into sub-images by using patch-based method and located the region of cracks by CBAM ResNet-50 model. The Recall reaches 0.95 on the test set. The second stage of our method uses the Attention U-Net model to get the accurate geometric edges of cracks based on results in the first stage. The IoU of the segmentation model implemented in this stage attains 0.48. In the third stage of the model, we remove the wrong-predicted isolated points in the predicted results through dilate operation and outlier elimination algorithm. The IoU of test set ascends to 0.70 after this stage. Ablation experiments are conducted to optimize the parameters and further promote the accuracy of the proposed method. The result shows that: (1) the best patch size of sub-images is 1024 × 1024. (2) the CBAM ResNet-50 and the Attention U-Net achieved the best results in the first and the second stage, respectively. (3) Pre-training the model of the first two stages can improve the IoU by 2.9%. In general, our method is of great significance for crack detection.