• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.2
• /
• pp.27-34
• /
• 2024

This study was conducted to examine the feasibility of using Fe-activated wood-derived biochar as a conductive filler for manufacturing cement-based strain sensor. To evaluate the compressive and electrical properties of cement composite with 3% Fe-activated biochar, three cubic specimens of size 50 x 50 x 50mm³ and three prismatic cement-based sensors of size 40 x 40 x 80mm³ were prepared respectively. The four-probe method of electrical resistance measurement was used for cement-based sensors. For cement-based sensors with FE-activated biochar, the conductive performance such as electrical resistance and impedance under different water content and repeated compression was investigated. Results showed that the fractional changes in the DC electrical resistivity of cement-based sensors increase with increasing time and the maximum fractional changes in the resistivity decrease with increasing the moisture contents during 900s. At moisture content of 7.5% range, the conductive performance of cement composite including 3% Fe-activated biochar as a conductive filler showed the most stable, while the strain detection ability tended to decrease somewhat as the repeated compressive stress increased between repeated compressive strain and fractional change in resistivity (FCR).

• Journal of Practical Engineering Education
• /
• v.13 no.2
• /
• pp.233-242
• /
• 2021

In this paper, as a liberal arts course for non-majors, we proposed a supervised learning artificial neural network parameter optimization method and a basic education method for activation function to design a basic artificial neural network subject curriculum. For this, a method of finding a parameter optimization solution in a spreadsheet without programming was applied. Through this training method, you can focus on the basic principles of artificial neural network operation and implementation. And, it is possible to increase the interest and educational effect of non-majors through the visualized data of the spreadsheet. The proposed contents consisted of artificial neurons with sigmoid and ReLU activation functions, supervised learning data generation, supervised learning artificial neural network configuration and parameter optimization, supervised learning artificial neural network implementation and performance analysis using spreadsheets, and education satisfaction analysis. In this paper, considering the optimization of negative parameters for the sigmoid neural network and the ReLU neuron artificial neural network, we propose a training method for the four performance analysis results on the parameter optimization of the artificial neural network, and conduct a training satisfaction analysis.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.292-294
• /
• 2002

본 연구에서는 영상에서 효과적인 에지 연결(edge linking)을 위하여 기존의 셀룰러 신경망 구조에서 두 가지 유형의 시냅스 구조를 고려한 활성화 특성을 제안한다 제시하는 모델에서 노드들간의 측면 연결에 의한 상호 작용은 노이즈에 의한 에지 및 영상에서 추출된 비최대점(non-maximum)의 에지를 억제할 문만 아니라, 특정 노드의 원형 이웃(circular neighborhood)으로 그려되는 특징들 간의 상호 연관도를 반영하여 에지의 연결 효과를 이를 수 있게 한다. 이러한 과정은 에지를 표현하는 벡터형식의 각 성분에 대한 활성화 특성으로부터 정형화된 에너지 함수로 모델링하고 이에 대한 최적화 과정으로써 구현될 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.15-16
• /
• 2009

OSEK/VDX 운영체제는 자동차 전자제어장치(ECU)를 위하여 OSEK/VDX에서 제안한 사양을 준수하는 실시간 운영체제이다. 본 연구는 OSEK/VDX 운영체제의 태스크관리 서비스들 중에서 태스크 활성화를 위한 서비스 함수를 설계하고 그 타당성을 제시한다.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2001.06a
• /
• pp.145-148
• /
• 2001

본 논문에서는 자기구성 신경회로망을 이용하여 3축 매니퓰레이터의 궤적제어기를 설계한다. 궤적 제어는 경유점을 정하고 각 경유점에 대한 역기구학을 적용하는 제어기로서 본 논문에서는 역기구학의 해를 자기구성 신경회로망을 통해 해결하는 제어기를 설계하고자 한 다. 또한 제어기에서의 은닉층의 활성화 함수는 가우 시안 함수를 사용하고, 은닉층의 파라미터는 오차를 기초로 하여 자동적으로 최적의 파라미터 값을 구함으로 서 유연한 궤적 제어가 되도록 한다.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.11a
• /
• pp.178-178
• /
• 2008

ZnO 바리스터는 인가되는 전압에 따라 저항이 변하는 전압 의존형 저항체이며 각종 전기 전자 정보통신용 제품에 정전기(ESD) 대책용 소자로 폭 넓게 사용되는 전자 세라믹스 부품이다. 특별히 Bi-based ZnO 바리스터는 다양한 상(phase)으로 구성되어 있으며 그 입계의 전기적 특성은 소량 첨가되는 dopant의 종류에 따라 다양하게 변하는 것으로 알려져 있다. 본 연구에서는 Bi-based ZnO 바리스터 (ZnO-$Bi_2O_3$, ZnO-$Bi_2O_3-Mn_3O_4$)에서 각종 유전함수$(Z^*,M^*,\varepsilon^*,Y^*,tan{\delta})$를 이용하여 입계의 주파수-온도에 대한 특성을 살펴 보았다. 일반적인 ZnO 바리스터 제조법으로 시편을 제작하여 78K~800K 온도 범위에서 각종 유전함수를 이용하여 복소 평면도(complex plane plot)와 주파수 응답도(frequency explicit plot)의 방법으로 defect level과 입계 특성(활성화 에너지, 정전용량, 저항, 입계 안정성 등)에 대하여 고찰하였다. ZnO-$Bi_2O_3$(ZB)계와 ZnO-$Bi_2O_3-Mn_3O_4$(ZBM)계 모두 상온 이하의 온도에서 $Zn_i$와 $V_o$의 결함이 나타났으며, 이들의 결함 준위는 각 유전함수에 따라 다소 차이가 났다. 입계 특성으로 ZB계는 이상구간(560~660K)을 전후로 1.15 eV $\rightarrow$ 1.49 eV의 활성화 에너지의 변화가 나타났지만, ZBM계는 이러한 현상이 나타나지 않았다. 또한 입계 전위 장벽의 온도 안정성에 대해서는 Cole-Cole model을 적용하여 분포 파라미터 (distribution parameter; $\alpha$)를 구하여 고찰하였다. ZB계의 입계 안정성은 온도에 따라 불안정해 졌지만, ZBM계는 안정하였다.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.611-616
• /
• 2023

ECG(electrocardiogram) is a test used to measure the rate and regularity of heartbeats, as well as the size and position of the chambers, the presence of any damage to the heart, and the cause of all heart diseases can be found. Because the ECG signal obtained using the ECG-KIT includes noise in the ECG signal, noise must be removed from the ECG signal to apply to the deep learning. In this paper, the noise of the ECG signal was removed using the digital IIR Butterworth low-pass filter. When the performance evaluation of the three activation functions, sigmoid(), ReLU(), and tanh() functions, was compared using the deep learning model of LSTM, it was confirmed that the activation function with the smallest error was the tanh() function. Also, When the performance evaluation and elapsed time were compared for LSTM and GRU models, it was confirmed that the GRU model was superior to the LSTM model.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.563-568
• /
• 2023

ECG(electrocardiogram) is used to measure the rate and regularity of heartbeats, as well as the size and position of the chambers, the presence of any damage to the heart, and the cause of all heart diseases can be found. Because the ECG signal obtained using the ECG-KIT includes noise in the ECG signal, noise must be removed from the ECG signal to apply to the deep learning. In this paper, Noise included in the ECG signal was removed by using a lowpass filter of the Digital FIR Hamming window function. When the performance evaluation of the three activation functions, sigmoid(), ReLU(), and tanh() functions, which was confirmed that the activation function with the smallest error was the tanh() function, the elapsed time was longer when the batch size was small than large. Also, it was confirmed that result of the performance evaluation for the GRU model was superior to that of the LSTM model.

• Journal of the Korean Society of Radiology
• /
• v.18 no.1
• /
• pp.21-26
• /
• 2024

The one of the famous deep learning models for object detection task is you only look once version 5 (YOLOv5) framework based on the one stage architecture. In addition, YOLOv5 model indicated high performance for accurate lesion detection using the bottleneck CSP layer and skip connection function. The purpose of this study was to evaluate the performance of YOLOv5 framework according to various hyperparameters in position emission tomogrpahy (PET) phantom images. The dataset was obtained from QIN PET segmentation challenge in 500 slices. We set the bounding box to generate ground truth dataset using labelImg software. The hyperparameters for network train were applied by changing optimization function (SDG, Adam, and AdamW), activation function (SiLU, LeakyRelu, Mish, and Hardwish), and YOLOv5 model size (nano, small, large, and xlarge). The intersection over union (IOU) method was used for performance evaluation. As a results, the condition of outstanding performance is to apply AdamW, Hardwish, and nano size for optimization function, activation function and model version, respectively. In conclusion, we confirmed the usefulness of YOLOv5 network for object detection performance in nuclear medicine images.

• Journal of the Korean Chemical Society
• /
• v.30 no.2
• /
• pp.166-171
• /
• 1986

Hydrogen atom transfer reactions, $RNH_2+CH_3{\to}RNH+CH_4\;where\;R=H\;and\;CH_3$, were studied by MINDO/3 MO method. It is expected that stable complex may exist between $CH_3\;and\;NH_3$. Transition state characteristics and activation parameters including activation entropies have been determined. In cases of absolute and activation entropies, results are found to agree in general with ab initio and experimental values, wherever comparisions are possible.