• Journal of the Semiconductor & Display Technology
• /
• v.10 no.3
• /
• pp.21-28
• /
• 2011

In recent years, owing to the development of the image processing technology, the research to build control system using a vision sensor is stimulated. However, a random time delay must be considered, because it works of a various time to get a result of an image processing in the system. It can be seen as an obstacle factor to a control of visual tracking in real system. In this paper, implementing two vision controllers each, first one is made up PID controller and the second one is consisted of a Smith Predictor, the possibility was shown to overcome a problem of a random time delay in a visual tracking system. A number of simulations and experiments were done to show the validity of this study.

• Journal of Information Processing Systems
• /
• v.13 no.6
• /
• pp.1554-1574
• /
• 2017

This paper introduces a new type of determining factor for Pseudo Random Strings (PRS). This classification depends upon a mathematical property called Finite Induction (FI). FI is similar to a Markov Model in that it presents a model of the sequence under consideration and determines the generating rules for this sequence. If these rules obey certain criteria, then we call the sequence generating these rules FI a PRS. We also consider the relationship of these kinds of PRS's to Good/deBruijn graphs and Linear Feedback Shift Registers (LFSR). We show that binary sequences from these special graphs have the FI property. We also show how such FI PRS's can be generated without consideration of the Hamiltonian cycles of the Good/deBruijn graphs. The FI PRS's also have maximum Shannon entropy, while sequences from LFSR's do not, nor are such sequences FI random.

• International journal of advanced smart convergence
• /
• v.10 no.2
• /
• pp.73-78
• /
• 2021

Recently, the number of people who visit the hospital due to diabetes is increasing. According to the Korean Diabetes Association, it is statistically indicated that one in seven adults aged 30 years or older in Korea suffers from diabetes, and it is expected to be more if the pre-diabetes, fasting blood sugar disorders, are combined. In the last study, the validity of Triglyceride and Cholesterol associated with diabetes was confirmed and analyzed using Random Forest. Random Forest has a disadvantage that as the amount of data increases, it uses more memory and slows down the speed. Therefore, in this paper, we compared and analyzed Random Forest and XGBoost, focusing on improvement of learning speed and prevention of memory waste, which are mainly dealt with in machine learning. Using XGBoost, the problem of slowing down and wasting memory was solved, and the accuracy of the diabetes recognition factor was further increased.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.5
• /
• pp.515-525
• /
• 2014

Recently, because of increasing in downpour and typhoon, which are caused by climate changes, those sedimentation disasters, such as landslide and debris flow, have become frequent. Those sedimentation disasters take place in natural slope. In order to predict debris flow damage range within wide area, the response model is more appropriate than numerical analysis. However, to make a prediction using Random Walk Model, the regional parameters is needed to be decided, since the regional environments conditions are not always same. This random Walk Model is a probability model with easy calculation method, and simplified slope factor. The objective of this study is to calculate the optimal parameters of Random Walk Model for Umyeon mountain in Seoul, where the large debris flow has occurred in 2011. Debris flow initiation zones and sedimentation zones were extracted through field survey, aerial photograph and visual reading of debris flow before and after its occurrence via LiDAR DEM.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.18 no.3
• /
• pp.323-334
• /
• 2008

According to the delivery condition, the breakage of a product occurs when it is delivered to the customers. Therefore product's makers evaluate the durability under the delivery process by accelerated life testing. In order to conduct this accelerated life testing accurately, it is very important to identify the acceleration-factor exactly between on-road and accelerated life test condition. In this paper, the acceleration-factor is identified by applying linear damage summation law, rain-flow cycle counting and Dirlik theory under the conditions of the random vibration. And approximated FEM model of the connecting-pipe to the compressor is developed for fatigue analysis. This model is finally verified by comparing the experiment results to the numerical analysis results.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.7
• /
• pp.2159-2166
• /
• 1996

The stochastic properties of variation in fatigue crack growth are important in reliability and stability of structures. In this study,the stochastic model for the variation of fatigue crack growth rate was proposed in consideration of nonhomogeneity of materials. For this model, experiments were ocnducted on 7075-T6 aluminum alloy under the constant stress intensity factor range. The variation of fatigue crack growth rate was expressed by random variables Z and r based on the variation of material coefficients C and m in the paris-Erodogan's equation. The distribution of fatigue life with respect to the stress intensity factor range was evaluated by the stochastic Markov chain model based on the Paris-Erdogan's equation. The merit of proposed model is that only a small number of test are required to determine this this function, and fatigue crack growth life is easily predicted at the given stress intensity factor range.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.4
• /
• pp.1340-1347
• /
• 2017

Utilization of Electric Vehicles (EVs) have been growing popularity in recent years due to increment in fuel price and lack of natural resources. Random unexpected charging by home EV charger in distribution is predicted in the future. The power quality problems such as fluctuation of power factor in a residential distribution network was explored with random EV chargers. This paper proposes a high-performance nonlinear sliding mode controller (SMC) for an EV charging system to compensate voltage distortions and to enhance the power factor against the unbalanced EV chargers. For the verification of the proposed scheme, MATLAB-Simulink simulations are performed on 22.9-kV grid. The results show that the proposed scheme can improve the power factor of a smart grid due to the EV chargers on the grid.

• Journal of information and communication convergence engineering
• /
• v.7 no.2
• /
• pp.209-214
• /
• 2009

The employment of the induction generator is preferable in the natural energy utilization by the minimum maintenance and the mechanical robustness, Another merit is also expected when it is connected to the power network system, because constant-voltage and constant frequency (CVCF) power generation is easily realized in spite of the variation of the rotor speed. However the induction generator needs much amount of the reactive power that reduces power factor in the primary side. The improvement of power factor in the primary side requires large VAR compensator, this point is solved, the merit of the induction machine as a main generator will become more established. This paper proposes a novel approach where the secondary is controlled by a PWM inverter not only to get CVCF power but also to improve the primary power factor. Basically the inverter is controlled so that the field current is supplied from the secondary side in this approach. The required capacity of the inverter is small, because only the slip power is controlled in the secondary side. In the experimental system where the sea wave torque simulator is used, the power factor is well improved by the microcomputer controlled PWM inverter.

• Structural Engineering and Mechanics
• /
• v.47 no.5
• /
• pp.643-660
• /
• 2013

In the present work, the capacity ratings of steel truss bridges have been carried out incorporating dynamic effect of moving vehicles and its accumulating effect as fatigue. Further, corrosion in the steel members has been taken into account to examine the rating factor. Dynamic effect has been considered in the rating procedure making use of impact factors obtained from simulation studies as well as from codal guidelines. A steel truss bridge has been considered to illustrate the approach. Two levels of capacity ratings- the upper load level capacity rating (called operating rating) and the lower load level capacity rating (called inventory rating) were found out using Load and Resistance Factor Design (LRFD) method and a proposal has been made which incorporates fatigue in the rating formula. Random nature of corrosion on the steel member has been taken into account in the rating by considering reduced member strength. Partial safety factor for each truss member has been obtained from the fatigue reliability index considering random variables on the fatigue parameters, traffic growth rate and accumulated number of stress cycle using appropriate probability density function. The bridge has been modeled using Finite Element software. Regressions of rating factor versus vehicle gross weight have been obtained. Results show that rating factor decreases when the impact factor other than those in the codal provisions are considered. The consideration of fatigue and member corrosion gives a lower value of rating factor compared to those when both the effects are ignored. In addition to this, the study reveals that rating factor decreases when the vehicle gross weight is increased.

• Computational Structural Engineering
• /
• v.2 no.4
• /
• pp.89-98
• /
• 1989

In order to take account of the statistical properties of random variables used in the structural analysis, the conventional approach usually adopts the safety factor based on past experiences for the qualitative assessment of structural safety problem. Recently, new approach based on the probabilistic concept has been applied to the assessment of structural safety in order to circumvent the difficulties of the conventional approach in choosing the appropriate safety factor. Thus, computer program called "Probabilistic finite element method" is developed by incorporating the probabilistic concept into the conventional matrix method in order to investigate the effects of the random variables on the final output of the structural analysis. From the comparison of some examples, it can be concluded that the PFEM developed in this study deals consistently with the uncertainty of random variables and provides the rational tool for the assessment of structural safety of plane frame.