• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.3
• /
• pp.380-387
• /
• 2010

For the fault diagnosis of three-phase induction motors, we propose a diagnosis algorithm based on mutual information and linear discriminant analysis (LDA). The experimental unit consists of machinery module for induction motor drive and data acquisition module to obtain the fault signal. As the first step for diagnosis procedure, DFT is performed to transform the acquired current signal into frequency domain. And then, frequency components are selected according to discriminate order calculated by mutual information As the next step, feature extraction is performed by LDA, and then diagnosis is evaluated by k-NN classifier. The results to verify the usability of the proposed algorithm showed better performance than various conventional methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.1
• /
• pp.454-459
• /
• 2017

In this study, a thermal system was designed using a 3-way valve and PTC attached to a solar module. This design could help solve the problem of rising fossil fuel costs caused by limited reserves and environmental problems resulting from fossil fuel use. The thermal system is a hot-air and heating control system composed of a temperature sensor part, mode setting part (for hot air and heating modes), supply part, and thermal system control part. The temperature sensor part has piping and an indoor temperature display, and the temperature setting part has multiple monitoring functions. The mode setting part switches between hot air and heating modes and can be used to set the temperature. The thermal system control part performs functions such as PTC control and temperature setting, PTC day and night and time selection, hot air and heating control, and three-way valve selection. The results verify that the system operates with stable response speeds of $680{\mu}s$ in the temperature sensor part, $700{\mu}s$ in the mode setting part, and $610{\mu}s$ in the thermal system control part.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.4B
• /
• pp.566-575
• /
• 2010

In recent, artificial immune system has become an important research direction in the anomaly detection of networks. The conventional artificial immune systems are usually based on the negative selection that is one of the computational models of self/nonself discrimination. A main problem with self and non-self discrimination is the determination of the frontier between self and non-self. It causes false positive and false negative which are wrong detections. Therefore, additional functions are needed in order to detect potential anomaly while identifying abnormal behavior from analogous symptoms. In this paper, we design novel network attack detection and response schemes based on artificial immune system, and evaluate the performance of the proposed schemes. We firstly generate detector set and design detection and response modules through adopting the interaction between dendritic cells and T-cells. With the sequence of buffer occupancy, a set of detectors is generated by negative selection. The detection module detects the network anomaly with a set of detectors and generates alarm signal to the response module. In order to reduce wrong detections, we also utilize the fuzzy number theory that infers the degree of threat. The degree of threat is calculated by monitoring the number of alarm signals and the intensity of alarm occurrence. The response module sends the control signal to attackers to limit the attack traffic.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.1
• /
• pp.18-24
• /
• 2007

If an error occurs in a product that contains a source of vibration, an abnormal noise vibration will occur. Recently a system that has been modified from the previous method of noise detection-a method of appraising the quality of manufactured automobile part by using human ears-is being implemented in the industries of automobile parts. This new system distinguishes the product's vibration signals by measuring and analyzing the signals. Following the recent trend, it has been concluded that the appraisal process of Window Regulator Module needed an improvement. Thus, a vibration monitoring system using LabVIEW, which measures and analyzes vibration signals from a sector gear's connected part by using an accelerometer, has been developed. By analyzing the characteristics of vibration signals of both inferior and superior goods, now the quality of the product can be evaluated much more accurately.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.801-806
• /
• 2008

Large quantity of bending deformation as well as rotating torque fluctuation at the balance shaft are main struggles during the operation in a high speed rotation and thereby, two issues should be cleared at the design process of balance shaft module. Since two issues are highly related with balance shaft itself and particularly much sensitive to the location of both supporting bearing and unbalance mass, the design strategy on balance shaft should be investigated at the aspect of controlling two critical issues at the early stage of balance shaft design. To tackle two main problems, the formulation of objective function that minimizes critical issues, both bending deformation as well as torque fluctuation, is suggested to derive the optimal information on balance shaft. Then, optimal informations are reviewed at the practical logics and the guideline at the selection of locations, both supporting bearing and unbalance mass, is addressed at the final chapter.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.7
• /
• pp.1553-1557
• /
• 2004

A real-time processing system for embedded hardware genetic algorithm is suggested. In order to operate basic module of genetic algorithm in parallel, such as selection, crossover, mutation and evaluation, dual processors based architecture is implemented. The system consists of two Xscale processors and two FPGA with evolvable hardware, which enables to process genetic algorithm efficiently by distributing the computational load of hardware genetic algorithm to each processors equally. The hardware genetic algorithm runs on Linux OS and the resulted chromosome is executed on evolvable hardware in FPGA. Furthermore, the suggested architecture can be extended easily for a couple of connected processors in serial, making it accelerate to compute a real-time hardware genetic algorithm. To investigate the effect of proposed approach, performance comparisons is experimented for an typical computation of genetic algorithm.

• Journal of Crop Science and Biotechnology
• /
• v.10 no.2
• /
• pp.73-78
• /
• 2007

The green revolution has significantly helped in increasing the food production. So far, various breeding methods have been exploited, besides them recombination DNA technology provides another approach for increasing the food production. By means of this technology the losses in food production incurred by various biotic and abiotic stresses can be effectively controlled. In most of the transgenic studies scientists have used antibiotic resistant genes as markers for easy selection of transformants but there are risks involved in use of GM foods. To make such foods safer and environment friendly we have discussed a novel strategy i.e. Cre-lox which involves site specific recombination. By means of Cre-lox the marker genes can be specifically removed once the selection of transformants is over. In addition, this strategy can be used to module the hybrid chromosomes, avoid gene silencing and incorporate single copy of a transgene for its higher expression.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.349-352
• /
• 2003

A personal low-frequency stimulator is a portable device to relax muscle pains of a person. The stimulator generates combined low-frequency pulses to be applied to pads attached to painful muscles. This paper reports a development of such device with its characteristic analyses. The major components of our stimulator are MCU, high-voltage generating circuit part, high-voltage switching circuit part, input switch part and display unit. High-voltage generating circuit is designed by using a boost converter circuit and allows user control of the output voltage. High-voltage switching circuit, controlled by MCU, generates output voltage to be applied to pads. Input switch part is composed of power supply, intensity selection, mode selection and memory. Display unit adopts a text LCD module to display modes, Intensity, output frequency and user set-up time. Our designed safety circuit, to protect human body from possible electric shock, slowly increases the output voltage to the selected output intensity. It continuously checks the output pulse shape and disable the output when dangerous pulses are detected. This paper also shows some experimental results.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.19 no.6
• /
• pp.857-870
• /
• 2017

The use of shield TBM is gradually increasing due to the urbanization of social infrastructures. Reliable estimation of advance rate is very important for accurate construction period and cost. For this purpose, it is required to develop the prediction model of advance rate that can consider the ground properties reasonably. Based on the database collected from field, statistical prediction procedure for field penetration index (FPI) was modularized in this study to calculate penetration rate of shield TBM. As output parameter, FPI was selected and various systems were included in this module such as, procedure of eliminating abnormal dataset, preprocessing of dataset and ridge regression with best subset selection. And it was finally validated by using field dataset.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.1
• /
• pp.50-55
• /
• 2021

Multi-path system is a system in which utilizes various networks simultaneously. It is expected that multi-path system can enhance communication speed, reliability, security of network. In this paper, we focus on path selection in multi-path system. To select optimal path, we propose deep reinforcement learning algorithm which is rewarded by the round-trip-time (RTT) of each networks. Unlike multi-armed bandit model, deep Q learning is applied to consider rapidly changing situations. Due to the delay of RTT data, we also suggest compensation algorithm of the delayed reward. Moreover, we implement testbed learning server to evaluate the performance of proposed algorithm. The learning server contains distributed database and tensorflow module to efficiently operate deep learning algorithm. By means of simulation, we showed that the proposed algorithm has better performance than lowest RTT about 20%.