• 천문학회보
• /
• 제46권2호
• /
• pp.46.3-47
• /
• 2021

Recently a multilayer spectral inversion (MLSI) model has been proposed to infer the physical parameters of plasmas in the solar chromosphere. The inversion solves a three-layer radiative transfer model using the strong absorption line profiles, H alpha and Ca II 8542 Å, taken by the Fast Imaging Solar Spectrograph (FISS). The model successfully provides the physical plasma parameters, such as source functions, Doppler velocities, and Doppler widths in the layers of the photosphere to the chromosphere. However, it is quite expensive to apply the MLSI to a huge number of line profiles. For example, the calculating time is an hour to several hours depending on the size of the scan raster. We apply deep neural network (DNN) to the inversion code to reduce the cost of calculating the physical parameters. We train the models using pairs of absorption line profiles from FISS and their 13 physical parameters (source functions, Doppler velocities, Doppler widths in the chromosphere, and the pre-determined parameters for the photosphere) calculated from the spectral inversion code for 49 scan rasters (~2,000,000 dataset) including quiet and active regions. We use fully connected dense layers for training the model. In addition, we utilize a skip connection to avoid a problem of vanishing gradients. We evaluate the model by comparing the pairs of absorption line profiles and their inverted physical parameters from other quiet and active regions. Our result shows that the deep learning model successfully reproduces physical parameter maps of a scan raster observation per second within 15% of mean absolute percentage error and the mean squared error of 0.3 to 0.003 depending on the parameters. Taking this advantage of high performance of the deep learning model, we plan to provide the physical parameter maps from the FISS observations to understand the chromospheric plasma conditions in various solar features.

• 한국전문물리치료학회지
• /
• 제24권2호
• /
• pp.27-36
• /
• 2017

Background: Research efforts to improve the pulmonary function of people with limited chest function have focused on the diaphragmatic ability to control breathing pattern. Real-time ultrasonography is appropriate to demonstrate diaphragmatic mechanism during breathing. Objective: The purpose of this study was to investigate the effects of diaphragmatic breathing training using real-time ultrasonographic imaging (RUSI) on the chest function of young females with limited chest mobility. Methods: Twenty-six subjects with limited chest mobility were randomly allocated to the experimental group (EG) and control group (CG) depending on the use of RUSI during diaphragmatic breathing training, with 13 subjects in each group. For both groups, diaphragmatic breathing training was performed for 30-min, including three 10-min sets with a 1-min rest interval. An extra option for the EG was the use of the RUSI during the training. Outcome measures comprised the diaphragmatic excursion range during quiet and deep breathing, pulmonary function (forced vital capacity; FVC, forced expiratory volume in 1-sec; FEV1, tidal volume; TV, and maximal voluntary ventilation; MVV), and chest circumferences at upper, middle, and lower levels. Results: The between-group comparison revealed that the diaphragmatic excursion range during deep breathing, FVC, and middle and lower chest circumferences were greater at post-test and that the changes between the pretest and post-test values were greater in the EG than in the CG (p<.05). In addition, the subjects in the EG showed increased post-test values for all the variables compared with the pretest values, except for TV and MVV (p<.05). In contrast, the subjects in the CG showed significant improvements for the diaphragmatic excursion range during quiet and deep breathings, FVC, FEV1, and middle and lower chest circumferences after the intervention (p<.05). Conclusion: These results indicate that using RUSI during diaphragmatic breathing training might be more beneficial for people with limited chest mobility than when diaphragmatic breathing training is used alone.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1033-1036
• /
• 2004

The study on speech recognition and understanding has been done for many years. In this paper, we propose a new type of recurrent neural network architecture for speech recognition, in which each output unit is connected to itself and is also fully connected to other output units and all hidden units [1]. Besides that, we also proposed the new architecture and the learning algorithm of recurrent neural network such as Backpropagation Through Time (BPTT, which well-suited. The aim of the study was to observe the difference of Arabic's alphabet like "alif" until "ya". The purpose of this research is to upgrade the people's knowledge and understanding on Arabic's alphabet or word by using Recurrent Neural Network (RNN) and Backpropagation Through Time (BPTT) learning algorithm. 4 speakers (a mixture of male and female) are trained in quiet environment. Neural network is well-known as a technique that has the ability to classified nonlinear problem. Today, lots of researches have been done in applying Neural Network towards the solution of speech recognition [2] such as Arabic. The Arabic language offers a number of challenges for speech recognition [3]. Even through positive results have been obtained from the continuous study, research on minimizing the error rate is still gaining lots attention. This research utilizes Recurrent Neural Network, one of Neural Network technique to observe the difference of alphabet "alif" until "ya".

• Journal of Astronomy and Space Sciences
• /
• 제32권4호
• /
• pp.305-310
• /
• 2015

In this study, we examined a morningside Pi2 pulsation, with a non-substorm signature, that occurred in very quiet geomagnetic conditions (Kp = 0) at 05:38 UT on December 8, 2012, using data obtained by Van Allen Probes A and B (VAP-A and VAP-B, respectively) and at a ground station. Using 1 sec resolution vector magnetic field data, we measured the X-component of the pulsation from the Abu Simbel ground station (L = 1.07, LT = UT +2 hr, where LT represents local time) in Egypt. At the time of the Pi2 event, Abu Simbel and VAP-A (L = 3.3) were in the morning sector (07:38 LT and 07:59 MLT, respectively, where MLT represents magnetic local time), and VAP-B was in the postmidnight sector (04:18 MLT and L = 5.7). VAP-A and VAP-B observed oscillations in the compressional magnetic field component (Bz), which were in close agreement with the X-component measurements of the Pi2 pulsation that were made at Abu Simbel. The oscillations observed by the satellites and on the ground were in phase. Thus, we concluded that the observed morningside Pi2 pulsation was caused by the cavity resonance mode rather than by ionospheric current systems.

• 대한기계학회논문집
• /
• 제18권10호
• /
• pp.2554-2566
• /
• 1994

Robot position/force control has been a difficult task owing to the interaction between a robot and an environment with a rather high stiffness. In addition to the dynamic instability, the interaction causes the following problem : 1) chattering at steady-state, 2) dynamic coupling effect of robot, and 3) performance degradation due to a titled environment. To solve the problem, the Time Delay Control(TDC), which has been known to be quiet robust to plant uncertainties and disturbances, has been applied. In conjunction to TDC, the following three ideas were also used : 1) To reduce the amplitude of the chattering at the steady state, a novel scheme was adopted to enhance the resolution type solution of A/D conversion for the force sensor. 2) To reduce the dynamic coupling, a trajectory type position command was tried on a comparative basis to the step command, as well as a more accurate mass matrix was used instead of the constant mass matrix. 3) And finally to improve the performance in the tilted environment, force derivatives instead of position derivatives were used in the TDC law. Computer simulations and experiments resulted in obvious improvements on the quality of the hybrid control, thereby clearly demonstrating the effectiveness of TDC with the proposed ideas.

• 무역상무연구
• /
• 제62권
• /
• pp.33-53
• /
• 2014

This study examines the seller's liability for defects in title of goods under SGA. If the contracting parties choose SGA as a governing law, they should pay attention to whether a contractual stipulation for defects in title of goods is a condition or a warranty. It is because SGA divides contractual terms into a condition and a warranty. And its effects regarding a breach of a condition or a warranty are different. Under SGA s 12(1) as a condition, in a contract of sale, the seller has a right to sell the goods at the time of contract, and in the case of an agreement to sell, he will have such a right at the time when the property is to pass. Under SGA s 12(2) as a warranty, there is an implied warranty that (a) the goods are free, and will remain free until the time when the property is to pass, from any charge or encumbrance and (b) the buyer will enjoy quiet possession of the goods as long as the buyer retains an interest in the goods. But the seller will not be liable if the third party unlawfully interferes with the buyer's possession.

• 대한방사선기술학회지:방사선기술과학
• /
• 제24권2호
• /
• pp.13-17
• /
• 2001

When most patients are diagnosed with the quiet progressed hepatoma which often would make the operation impossible, the Interventional Radiology hepatic artery embolization is an extremely useful method for such patients. An existence of the malfunction is evaluated by gaining a portal vein image as a delayed phase image after injecting a contrast media into the superior mesenteric artery. However, it is difficult to make a definite judgement due to the extended exposure time with the peristalsis and the intestine gas obstructing the sharpness of the image when the Patient exposure time increases and due to the increased usage of contrast media and its side effect. The portal vein can be evaluated by obtaining the MIP image after reconstructing a 3-dimensional personal computer setting using the 2-dimensional from an enhancement abdomen CT image that is almost a requisite in operation to a hepatoma patient. Such method nay prevent a decrease in the quality of image based upon the time delay and intestine gas; also, because the patient exposure dose and contrast media usage may be reduced, it is a new, valuable way to decide the operational matter of hepatic artery embolization on a pre-angiography.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.620-628
• /
• 2005

Spatial control of sound is essential to deliver better sound to the listener's position in space. As it can be experienced in many listening environments, the quality of sound can not be manifested over every position in a hall. This motivates us to control sound in a region we select. The primary focus of the developed method has to do with the brightness and contrast of acoustic image in space. In particular, the acoustic brightness control seeks a way to increase loudness of sound over a chosen area, and the contrast control aims to enhance loudness difference between two neighboring regions. This enables us to make two different kinds of zone - the zone of quiet and the zone of loud sound - at the same time. The other perspective of this study is on the direction of sound. It is shown that we can control the direction of perceived sound source by focusing acoustic energy in wavenumber domain. To begin with, the proposed approaches are formulated for pure-tone case. Then the control methods are extended to a more general case, where the excitation signal has broadband spectrum. In order to control the broadband signal in time domain, an inverse filter design problem is defined and solved in frequency domain. Numerical and experimental results obtained in various conditions certainly validate that the acoustic brightness, acoustic contrast, direction of wave front can be manipulated for some finite region in space and time.

• KIEAE Journal
• /
• 제1권2호
• /
• pp.13-19
• /
• 2001

Recently, With the increase of income level, many people desire to have agreeable and quiet surrounding. However, the extent and time of the exposure to construction noise and vibration is increasing, which is a factor that ruins the quietness of the surrounding and increasingly raises interests in its effects. In this point, this study attempts to survey the characteristics of construction vibration by SD(Sand drain) method in construction field. In addition, the result of study is presumed to be usable as the fundamental materials for an efficient counter-plan in case of the public resentment in relations to the construction vibration occurring henceforth.

• 대한기계학회논문집A
• /
• 제22권4호
• /
• pp.879-886
• /
• 1998

It is very important to understand the vibration and noise characteristics of a structure to developed quiet machines and lessen their noise. In this paper, the vibration and sound radiation characteristics of a simple and a bar-stiffened plate have been investigated using numerical and experimental techniques. In numerical process, FEM analysis has been performed for the vibration level ; the time-space squared and averaged velocity and BEM analysis for sound radiation parameters ; sound power and radiation efficiency. In experimental process, FFT signal processing method has been used. While a power from an exiciter is applied to the structure by using a point contact, sound intensity and vibration level has been measured. Based on these two data, the radiation efficiency has been calculated. Results show that the radiation efficiency for the stiffened structure increases compared to the simple plate, due to the extra edges provided by the stiffener.