• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.29-37
• /
• 2017

The range of problems that can be handled by the activation of big data and the development of hardware has been rapidly expanded and machine learning such as deep learning has become a very versatile technology. In this paper, mnist data set is used as experimental data, and the Cross Entropy function is used as a loss model for evaluating the efficiency of machine learning, and the value of the loss function in the steepest descent method is We applied the Gradient Descent Optimize algorithm to minimize and updated weight and bias via backpropagation. In this way we analyze optimal reliability value corresponding to the number of exercises and optimal reliability value without overfitting. And comparing the overfitting time according to the number of data changes based on the number of training times, when the training frequency was 1110 times, we obtained the result of 92%, which is the optimal reliability value without overfitting.

• Structural Engineering and Mechanics
• /
• v.65 no.1
• /
• pp.111-120
• /
• 2018

This paper presents a new and simple solution for determining the natural frequencies of framed tube combined with shear-walls and tube-in-tube systems. The novelty of the presented approach is based on the bending moment function approximation instead of the mode shape function approximation. This novelty makes the presented solution very simpler and very shorter in the mathematical calculations process. The shear stiffness, flexural stiffness and mass per unit length of the structure are variable along the height. The effect of the structure weight on its natural frequencies is considered using a variable axial force. The effects of shear lag phenomena has been investigated on the natural frequencies of the structure. The whole structure is modeled by an equivalent non-prismatic shear-flexural cantilever beam under variable axial forces. The governing differential equation of motion is converted into a system of linear algebraic equations and the natural frequencies are calculated by determining a non-trivial solution for the system of equations. The accuracy of the proposed method is verified through several numerical examples and the results are compared with the literature.

• Biomedical Science Letters
• /
• v.4 no.2
• /
• pp.87-101
• /
• 1998

There is considerable interests in the effect of regular, vigorous exercise, in particular weight training as a possible measure in improving myocardial function. The present investigation aimed to examine possible effect of a long-term weight training program on the heart in aging rats. Male rats aged 3-,10-, and 20-month-old were divided at random into a control (sedentary) and the exercise group. The training group was exercised for 5 days a week by 1 RM of ten times with weight training apparatus. This investigation examined the changes of the heart muscle relative to histological, ultrastructural, cytochemical, and stereological studies in rats. Quantity of lipofuscin pigments was clearly increased in the weight training group of 15-month-old rats, and mitochondrial degeneration, vacuolization, and interstitial proliferation were observed as well. In the weight training group of 25-month-old rats the frequency as separated intercalated discs, fraying myofibrils and hypercontraction band increased in number compared with the same 15-mon1h-old group. From the experimental result of glucose-6-phosphatase activity, the enzyme activities decreased in the weight training group of 15-month-old rats, and more decreased in the same 25-month-old group. In stereological study, both 15- and 25-month-old training groups, mitochondrial and myofibrilar volume densities significantly decreased, whereas interstitial volume density significantly increased. From the experimental results obtained in the present study, it is suggested that long-term weight training exercise do not cause any significantly qualitative and quantitative ultrastructural change of the heart muscle in the young. On the contrary, long-term weight training exercise stress may actually induce degenerative changes in the heart muscle in the old age.

• Journal of Korean Academy of Nursing
• /
• v.13 no.3
• /
• pp.1-33
• /
• 1983

The role and function of nursing care nowadays, tend to acquire sophisicated technology because specialization has expended due to increase of the medical population and the improvement of national health standards. To implement nursing care independently as a professional nurse, the apprehension of specific knowledge and skill should be acquired during basic nursing education. So it is important for nursing education not only to include theory and actual techniques, but also to strengthen the practical training in the actual clinical setting. This study was carried out with the following objectives; 1. To survey the detailed content and frequency of actual nursing students display during their clinical training. 2. To investigate the detailed content and frequency of actual nursing behavior which students display in each clinical a area. 3. To identify the motive for selection of nursing as their major and to determine the degree of self confidence, extent of knowledge and recognition of nursing responsibility. 4. To observe the relationship between actual nursing behavior and each of the following; 1) Motive for selecting nursing as a major 2) Self confidence 3) Knowledge of nursing care 4) Recognition of nursing responsibility The conclusions of this study were as follows; 1. Among the detailed nursing behavior which junior nursing college students carry out in clinical training; taking respiration's showed the highest frequency, and taking body temperatures, blood pressures, and pulses and making beds were next in frequency in this order. 2. In detailed nursing behaviors according to clinical area; taking vital signs showed the highest frequency in the emergency room, pediatric ward, orthopedic ward, general surgical ward and internal medicine ward. However, in the operating room, assisting with endotracheal tube insertion and sterile techniques were showen to have the highest frequencies. In nursery, umbilical cord care and the measurement of body weight were the highest in frequency In neurosurgical ward, the measurement of vital signs, changing position and tracheostomy care were the highest in frequency. In obstetric and gynecological ward and in the delivery room, checking duration, intensity and frequency of contractions was the highest in frequency. 3. In regard to the motive for majoring in nursing, the aptitude and interest of the student had the highest percentage(32.86%), and self-confidence in nursing activities (M=3.36), knowledge in nursing activities.(M=3. 09), and the recognition of the nursing activity (M= 3.76) wire in the middle range. 4. When the detailed nursing behaviors were compared with motive, self confidence, knowledge and recognition, it was found that when the nursing behavior was difficult and regarding much endeavor although the motive was high, the frequency of the nursing behavior was rather low. But in the cases in which there was much self confidence and a high level of skill was required, nursing behavior was carried more frequently. When there was muck self confidence and skill was not required, the frequency of nursing behavior was rather low. In the cases of a high level of knowledge, the frequency of nursing behavior was low and when recognition for nursing behavior was given the frequency of nursing behavior was low.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.1
• /
• pp.1-10
• /
• 2005

This paper presents a wideband LPA (local polynomial approximation) beamforming algorithm that is appropriate for wideband moving sources. The Proposed wideband LPA algorithm adopts STMV (steered minimum variance) method that utilizes a steered covariance matrix obtained from multiple frequency components in one data snapshot, instead of multiple data snapshots in one frequency bin. The wideband LPA cost function is formed using STMV weight vector. The Proposed algorithm searches for the instantaneous DOA and angular velocity that maximize the wideband LPA cost function. resulting in a higher resolution performance than that of a DS LPA beamforming algorithm. Several simulations using artificial data and sea trial data are used to demonstrate the performance of the Proposed algorithm.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.57-65
• /
• 2023

A vibration response-based detection system was used to investigate the adhesive areas of single-lap joints using a nonlinear transformation approach for deep learning. In industry or engineering fields, it is difficult to know the condition of an invisible part within a structure that cannot easily be disassembled and the conditions of adhesive areas of adhesively bonded structures. To address these issues, a detection method was devised that uses nonlinear transformation to determine the adhesive areas of various single-lap-jointed specimens from the vibration response of the reference specimen. In this study, a frequency response function with nonlinear transformation was employed to identify the vibration characteristics, and a virtual spectrogram was used for classification in convolutional neural network based deep learning. Moreover, a vibration experiment, an analytical solution, and a finite-element analysis were performed to verify the developed method with aluminum, carbon fiber composite, and ultra-high-molecular-weight polyethylene specimens.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.11
• /
• pp.1578-1583
• /
• 2013

In this paper, stochastic optimization algorithms of PSO (Particle Swarm Optimization) and APSO (Adaptive Particle Swam Optimization) are studied and compared. It is revealed that the APSO provides faster convergence and better search efficiency than the conventional PSO when they are adopted to find the global minimum of a two-dimensional function. The advantages of the APSO comes from the ability to control the inertia weight, and acceleration coefficients. To verify that the APSO is working better than the standard PSO, the design of a 10GHz microstrip patch as one of the elements of a high frequency array antenna is taken as a test-case and shows the optimized result with 5 iterations in the APSO and 28 iterations in th PSO.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.391-394
• /
• 1997

A H$_{\infty}$ control techniques with roll eccentricity filter is proposed to alleviate the effect of entry thickness variation and roll eccentricity occurred in rolling stand itself of tandem cold mills. A robust controller to the disturbances is designed using H$_{\infty}$ control techniques, which can reflect the input direction of disturbances and knowledge of disturbance spectrum in the frequency domain. And, non-standard H$_{\infty}$ control problem caused by selection of weight function having poles on j.omega. axis is discussed. The evaluation for the resultant controller composed by H$_{\infty}$ synthesis is done through computer simulations. The effectiveness of the proposed method is compared to those of the conventional LQ synthesis method and a feedforward controller against roll eccentricity, which was already studied.ied.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.6
• /
• pp.691-700
• /
• 2007

This paper proposes a simple but effective design method of PID control using a numerical optimization method. In order to achieve both stability and performance, gain and phase margins and performance indices of step response directly compose of the cost function. Hence, the proposed approach is a multiobjective optimization problem. The main effectiveness of this approach results from the strong capability of the used optimization method. A one-dimensional example concerning gain margin illustrates the practical applicability of the optimization method. The present approach has many degrees of freedom in controller design by only adjusting related weight constants. The attained PID controller is compared with Wang#s and Ho#s methods, IAE, and ISE for a high-order process, and the simulation result for various design targets shows that the proposed approach achieves desired time-domain performance with a guarantee of frequency-domain stability.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.23 no.3
• /
• pp.266-272
• /
• 2014

This study proposes a closed-loop shaping control method with $H_{\infty}$ optimization for optical image stabilization (OIS) in mobile phone cameras. The image stabilizer is composed of a horizontal stage constrained by ball bearings and actuated by the magnetic force from voice coil motors. The displacement of the stage is measured by Hall effect sensors. From the OIS frequency response experiment, the transfer function models of the stage and Hall effect sensor were identified. The weight functions were determined considering the tracking performance, noise attenuation, and stability with considerable margins. The $H_{\infty}$ optimal controller was executed using closed-loop shaping and limiting the controller order, which should be less than 6 for real-time implementation. The control algorithm was verified experimentally and proved to operate as designed.