• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.511-516
• /
• 2002

The neural networks may have problem such that the amount of calculation for the network learning goes too big according to the dimension of the dimension. To overcome this problem, the wavelet neural networks(WNN) which use the orthogonal basis function in the hidden node are proposed. One can compose wavelet functions as activation functions in the WNN by determining the scale and center of wavelet function. In this paper, when we compose the WNN using wavelet functions, we set a single scale function as a node function together. We intend that one scale function approximates the target function roughly, the other wavelet functions approximate it finely During the determination of the parameters, the wavelet functions can be determined by the global search for solutions suitable for the suggested problem using the genetic algorithm and finally, we use the back-propagation algorithm in the learning of the weights.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.1
• /
• pp.116-124
• /
• 2009

This paper proposes a new Artificial-Intelligent(AI) Power Quality(PQ) diagnosis algorithm using Discrete Wavelet Transform(DWT), Fast Fourier Transform(FFT), Root-Mean-Square(RMS) value. The developed algorithm is able to detect and classify the PQ problems such as the transient, the voltage sag, the voltage swell, the voltage interruption and the total harmonics distortion. The 15.36[kHz] sampling frequency is used to measure the voltages in a power system. The measured signals are used for DWT, FFT, RMS calculation. For AI diagnosis of the PQ problems, a simple multi-layered Artificial Neural Network(ANN) with the back-propagation algorithm is adopted, programmed in C++ and tested in PSIM simulation studies. Finally, the algorithm, which is installed in MP PQ+256 with TI DSP320C6713, is proved to diagnose the PQ problems efficiently.

• Asian Oncology Nursing
• /
• v.5 no.1
• /
• pp.3-10
• /
• 2005

Back ground: Nurses now occupy one third of all hospital human resources. Therefore, efficient management of nursing manpower is getting more important. While it is very clear that nursing workload requirement analysis and patient severity classification should be done first for the efficient allocation of nursing workforce, these processes have been conducted manually with ad hoc rule. Purposes: This study was tried to make a predict model for patient classification according to nursing need. We tried to find the easier and faster method to classify nursing patients that can help efficient management of nursing manpower. Methods: The nursing patient classifications data of the hospitalized cancer patients in one of the biggest cancer center in Korea during 2003.1.1-2003.12.31 were assessed by trained nurses. This study developed a prediction model and analyzing nursing needs by data mining techniques. Patients were classified by three different data mining techniques, (Logistic regression, Decision tree and Neural network) and the results were assessed. Results: The data set was created using 165,073 records of 2,228 patients classification database. Main explaining variables were as follows in 3 different data mining techniques. 1) Logistic regression : age, month and section. 2) Decision tree : section, month, age and tumor. 3) Neural network : section, diagnosis, age, sex, metastasis, hospital days and month. Among these three techniques, neural network showed the best prediction power in ROC curve verification. As the result of the patient classification prediction model developed by neural network based on nurse needs, the prediction accuracy was 84.06%. Conclusion: The patient classification prediction model was developed and tested in this study using real patients data. The result can be employed for more accurate calculation of required nursing staff and effective use of labor force.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.2
• /
• pp.1-9
• /
• 2003

In the present study, a 3-D Parallel DSMC method in developed on unstructured meshes for the efficient simulation of rarefied gas flows. Particle tracing between cells in achieved based on a linear shape function extended to three dimensions. For high parallel efficiency, successive domain decomposition is applied to achieve load balancing between processors by accounting for the number of particles. A particle weighting technique is also adopted to handle flows containing gases of significantly dirrerent number densities in the same flow domain. Application is made for flow past a 3-D delta wing and the result is compared with that from experiment and other calculation. Flow around a rocket payload at 100km altitude is also solved and the effect of plume back flow from the nozzle in studied.

• Geomechanics and Engineering
• /
• v.14 no.5
• /
• pp.467-477
• /
• 2018

Estimation of hydraulic parameters is a critical step during design of foundation dewatering works. When many piles are installed in an aquifer, estimation of the hydraulic conductivity should consider the blocking of groundwater seepage by the piles. Based on field observations during a dewatering project in Shanghai, hydraulic conductivities are back-calculated using a numerical model considering the actual position of each pile. However, it is difficult to apply the aforementioned model directly in field due to requirement to input each pile geometry into the model. To develop a simple numerical model and find the optimal hydraulic conductivity, three scenarios are examined, in which the soil mass containing the piles is considered to be a uniform porous media. In these three scenarios, different sub-regions with different hydraulic conductivities, based on either automatic inverted calculation, or on effective medium theory (EMT), are established. The results indicate that the error, in the case which determines the hydraulic conductivity based on EMT, is less than that determined in the automatic inversion case. With the application of EMT, only the hydraulic conductivity of the soil outside the pit should be inverted. The soil inside the pit with its piles is divided into sub-regions with different hydraulic conductivities, and the hydraulic conductivity is calculated according to the volume ratio of the piles. Thus, the use of EMT in numerical modelling makes it easier to consider the effect of piles installed in an aquifer.

• Geomechanics and Engineering
• /
• v.12 no.1
• /
• pp.139-160
• /
• 2017

The bearing mechanism of tunnel-type anchorage (TTA) for suspension bridges is studied. Model tests are conducted using different shapes of plug bodies, which are circular column shape and circular truncated cone shape. The results show that the plug body of the latter shape possesses much larger bearing capacity, namely 4.48 times at elastic deformation stage and 4.54 times at failure stage compared to the former shape. Numerical simulation is then conducted to understand the mechanical and structural responses of plug body and surrounding rock mass. The mechanical parameters of the surrounding rock mass are firstly back-analyzed based on the monitoring data. The calculation laws of deformation and equivalent plastic strain show that the numerical simulation results are rational and provide subsequent mechanism analysis with an established basis. Afterwards, the bearing mechanism of TTA is studied. It is concluded that the plug body of circular truncated cone shape is able to take advantage of the material strength of the surrounding rock mass, which greatly enhances its bearing capacity. The ultimate bearing capacity of TTA, therefore, is concluded to be determined by the material strength of surrounding rock mass. Finally, recommendations for TTA design are proposed and discussed.

• The KIPS Transactions:PartA
• /
• v.16A no.6
• /
• pp.473-480
• /
• 2009

This paper is to enhance a color image running in the PXA255 ARM processor based on embedded linux environments. Retinex is one of the representative algorithm for image enhancement in the previous research. However, retinex is not suitable the run on the embedded system because of its long processing time. So, we proposed the image enhancement algorithm for embedded system, with less quantity of operation and the effect equivalent to retinex. To achieve this goal, we propose and implement the image enhancement algorithm, which utilizes the image formation model and gamma correction to be effective in a back-light and dark image. The proposed algorithm converts the color space from RGB to HSV, and then V and S channels are processed. In order to optimize the proposed method in the PXA255 ARM processor, quantity of calculation is reduced. The performance of the proposed algorithm was evaluated through qualitative method and quantitative method. The results show that brightness and contrast are improved with less quantity of operation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.9 no.1
• /
• pp.53-62
• /
• 1989

In order to determine the form and the length of saline wedge, it is necessary to evaluate interfacial friction factor. Hetherto one dimensional two-layer flow model which assumed pressure as the hydrostatic pressure distribution has been well used to the calculation of saline wedge form, it just then stands in need of relevant interfacial friction factor. For example, in the case where we calculate back to interfacial friction factor out of saline wedge form obtained at a laboratory open channel with comparatively narrow width, it is needed to correct the side-wall effect of a channel, if generally negligible in the river. In this study, we confirmed the influence of a side-wall upon the lateral velocity distributions at laboratory channel and then examined in detail the value of interfacial friction factor in the case where it was corrected by the side-wall effect and not corrected. And then we make clear the influence of a side-wall upon the arrested saline wedge and interfacial faction factor from these results.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.24 no.3
• /
• pp.213-218
• /
• 2014

CFD flow simulation of vehicles with open sunroof and passenger window help the automotive OEM(original equipment manufacturer) to identify the low frequency noise levels in the cabin. The lock-in and lock-off phenomena observed in the experimental studies of sunroof buffeting is well predicted by CFD speed sweep calculations over the operating speed range of the vehicle. The trend of the shear layer oscillation frequency with vehicle speed is also well predicted. The peak SPL from the CFD calculation has a good compromise with the experimental value after incorporating the real world effects into the CFD model by means of artificial compressibility and damping correction. The entire process right from modeling to flow analysis as well as acoustic analysis has been performed within the single environment i.e., STAR-CCM+.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.30 no.6 s.249
• /
• pp.553-559
• /
• 2006

Most diesel injector, which is currently used in high-pressure common rail fuel injection system of diesel engine, is driven by the solenoid coil energy for its needle movement. The main disadvantage of this solenoid-driven injector is a high power consumption, high power loss through solenoid coil and relatively fixed needle response's problem. In this study, a prototype piezo-driven injector, as a new injector mechanism driven by piezoelectric energy based on the concept of inverse piezo-electric effect, has been designed and fabricated to know the effect of piezo-driven injection processes on the diesel spray structure and internal nozzle flow. Firstly we investigated the spray characteristics in a constant volume chamber pressurized by nitrogen gas using the back diffusion light illumination method for high-speed temporal photography and also analyzed the inside nozzle flow by a fully transient simulation with cavitation model using VOF(volume of fraction) method. The numerical calculation has been performed to simulate the cavitating flow of 3-dimensional real size single hole nozzle along the injection duration. Results were compared between a conventional solenoid-driven injector and piezo-driven injector, both equipped with the same micro-sac multi-hole injection nozzle. The experimental results show that the piezo-driven injector has short injection delay and a faster spray development and produces higher injection velocity than the solenoid-driven injector. And the predicted simulation results with the degree of cavitation's generation inside nozzle for faster needle response In a piezo-driven injector were reflected to spray development in agreement with the experimental spray images.