• Korean Chemical Engineering Research
• /
• v.59 no.1
• /
• pp.6-10
• /
• 2021

The shorting resistance (SR) of the PEMFC(Proton Exchange Membrane Fuel Cell) polymer membrane is an important indicator of the durability of the membrane. When SR decreases, shorting current (SC) increases, reducing durability and performance. When SR becomes less than about 0.1 kΩ·㎠, shorting occurs, the temperature rises rapidly, and MEA(Membrane Electrode Assembly) is burned to end stack operation. In order to prevent shorting, we need to control the SR, so the conditions affecting the SR were studied. There were differences in the SR measurement methods, and the SR measurement method, which improved the DOE(Department of Energy) and NEDO(New Energy and Industrial Technology Development Organization) method, was presented. It was confirmed that the SR decreases as the relative humidity, temperature and cell compression pressure increase. In the final stage of the accelerated durability evaluation process of the polymer membrane, SR rapidly decreased to less than 0.1 kΩ·㎠, and the hydrogen permeability became higher than 15 mA/㎠. After dismantling the MEA, SEM(Scanning Electron Microscope) analysis showed that a lot of platinum was distributed inside the membrane.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.1
• /
• pp.79-89
• /
• 2022

Soil moisture is significantly related to crop growth and plays an important role in irrigation management. To predict soil moisture, various process-based model has been developed and used in the world. Various models (Land surface model) may have different performance depending on the model parameters and structures that causes the different model output for the same modeling condition. In this study, the three land surface models (Noah Land Surface Model, Soil Water Atmosphere Plant, Community Land Model) were used to compare the model performance (soil moisture prediction) and develop the multi-model simulation. At first, the genetic algorithm was used to estimate the optimal soil parameters for each model, and the parameters were used to predict soil moisture in the study area. Then, we used the multi-model approach based on Bayesian model averaging (BMA). The results derived from this approach showed a better match to the measurements than the results from the original single land surface model. In addition, identifying the strengths and weaknesses of the single model and utilizing multi-model methods can help to increase the accuracy of soil moisture prediction.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1363-1372
• /
• 2021

Red tide, defined as the major reproduction of harmful birds, has the characteristics of being generated and diffused in a wide area. This has limitations in detection only with the existing investigation method. Therefore, in this study, red tide was detected using a remote sensing technique. In addition, it was intended to increase the accuracy of detection by using optical characteristics, not just the concentration of chlorophyll. Red tide mainly occurs on the southern coast where sea signals are complex, and the main red tide control species on the southern coast is Cochlodinium polykirkoides. Therefore, it was intended to secure objectivity by reflecting features that could not be found depending on the researcher's observation and experience, not limited to visual judgment using machine learning techniques. In this study, support background machines and random forest were used among machine learning models, and as a result of calculating accuracy as performance evaluation indicators of the two models, the accuracy was 85.7% and 80.2%, respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.69-76
• /
• 2022

This paper proposes a method of refining and processing time-series data using Medical Information Mart for Intensive Care (MIMIC-IV) v2.0 data. In addition, the significance of the processing method was validated through a machine learning-based pressure ulcer early warning system using a dataset processed based on the proposed method. The implemented system alerts medical staff in advance 12 and 24 hours before a lesion occurs. In conjunction with the Electronic Medical Record (EMR) system, it informs the medical staff of the risk of a patient's pressure ulcer development in real-time to support a clinical decision, and further, it enables the efficient allocation of medical resources. Among several machine learning models, the GRU model showed the best performance with AUROC of 0.831 for 12 hours and 0.822 for 24 hours.

• Journal of the Society of Disaster Information
• /
• v.19 no.3
• /
• pp.689-697
• /
• 2023

Purpose: The purpose of this study is to automatically extract information on underground facilities using a general-purpose smartphone in the process of applying the mini-trenching method. Method: Data sets for image learning were collected under various conditions such as day and night, height, and angle, and the object detection algorithm used the R-CNN algorithm. Result: As a result of the study, F1-Score was applied as a performance evaluation index that can consider the average of accurate predictions and reproduction rates at the same time, and F1-Score was 0.76. Conclusion: The results of this study showed that it was possible to extract information on underground buried materials based on smartphones, but it is necessary to improve the precision and accuracy of the algorithm through additional securing of learning data and on-site demonstration.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.201-201
• /
• 2023

일반적으로 하천의 수질은 산업화, 인구증가 등으로 인해 여러 종류의 오염물질이 유입되어 악화된다. 수질 악화의 대표적인 현상은 부영양화이며 이를 일으키는 주요 원인 물질은 통상 영양염류라고 말하는 질소와 인으로 알려져 있다. T-N이 다량 수계로 유입되면 식물성 플랑크톤 등이 대량 번식하여 녹조 현상등 수질 악화를 발생시켜 관리가 필요하다. 현재 많은 수자원 관리 부서에서 모니터링 포인트를 설정하여 수질 변화를 관찰하고 있다. 기존의 T-N 분석방법은 (1) 자외선 흡광광도법 (2) 카드뮴 환원법 (3) 환원증류-킬달법등이 있다. 그러나 이러한 방법들은 실험실 기반의 정량적 분석으로 시간과 비용이 크게 소요되어 발생하는 문제에 대해 초기대응을 하기 힘들다. 따라서 T-N을 효과적으로 측정할 수 있는 방법이 필요하다. 국내에서는 수질자료를 통한 연관된 수질 인자를 찾아내어 머신러닝 알고리즘을 활용해 Chl-a 농도를 추정한 연구사례가 있다. 국외에서는 TN과 센서 측정 지표 간의 물리적, 화학적 관계를 기반으로 센서 감지의 적시성과 지능형 알고리즘의 정확도를 결합하여 실시간 총질소(TN) 측정 방법 연구 사례가 있다. 따라서 본 연구에서는 머신러닝을 활용하여 국내에 적합한 T-N 예측 모델을 만들고자한다. 본 연구에서는 센서기반으로 측정가능한 수질항목들과 T-N의 상관성 분석을 통해 주요 수질인자를 도출하였다. 도출된 인자와 Python 기반의 머신러닝을 활용하여 T-N을 추정하였다. 그 후, T-N 추정값과 실측값을 비교하여 머신러닝 성능을 평가하고 실제 적용 가능성에 대해서 검증하였다. 본 연구는 기존 T-N 측정에 소모되는 시간과 비용의 감소에 기여하고 이를 통해 앞으로 더 정확한 수질 예측이 가능해질 것으로 기대된다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.234-240
• /
• 2022

This paper proposes a customized AI exercise recommendation service for balancing the relative amount of exercise according to the working environment by each occupation. WISDM database is collected by using acceleration and gyro sensors, and is a dataset that classifies physical activities into 18 categories. Our system recommends a adaptive exercise using the analyzed activity type after classifying 18 physical activities into 3 physical activities types such as whole body, upper body and lower body. 1 Dimensional convolutional neural network is used for classifying a physical activity in this paper. Proposed model is composed of a convolution blocks in which 1D convolution layers with a various sized kernel are connected in parallel. Convolution blocks can extract a detailed local features of input pattern effectively that can be extracted from deep neural network models, as applying multi 1D convolution layers to input pattern. To evaluate performance of the proposed neural network model, as a result of comparing the previous recurrent neural network, our method showed a remarkable 98.4% accuracy.

• Smart Media Journal
• /
• v.12 no.5
• /
• pp.28-35
• /
• 2023

In this study, we propose a new light-weight model RoutingConvNet with fewer parameters to improve the applicability and practicality of speech emotion recognition. To reduce the number of learnable parameters, the proposed model connects bidirectional MFCCs on a channel-by-channel basis to learn long-term emotion dependence and extract contextual features. A light-weight deep CNN is constructed for low-level feature extraction, and self-attention is used to obtain information about channel and spatial signals in speech signals. In addition, we apply dynamic routing to improve the accuracy and construct a model that is robust to feature variations. The proposed model shows parameter reduction and accuracy improvement in the overall experiments of speech emotion datasets (EMO-DB, RAVDESS, and IEMOCAP), achieving 87.86%, 83.44%, and 66.06% accuracy respectively with about 156,000 parameters. In this study, we proposed a metric to calculate the trade-off between the number of parameters and accuracy for performance evaluation against light-weight.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.67-77
• /
• 2023

This study represents an innovative research conducted in the smart farm environment, developing a deep learning-based disease and pest detection model and applying it to the Intelligent Internet of Things (IoT) platform to explore new possibilities in the implementation of digital agricultural environments. The core of the research was the integration of the latest ImageNet models such as Pseudo-Labeling, RegNet, EfficientNet, and preprocessing methods to detect various diseases and pests in complex agricultural environments with high accuracy. To this end, ensemble learning techniques were applied to maximize the accuracy and stability of the model, and the model was evaluated using various performance indicators such as mean Average Precision (mAP), precision, recall, accuracy, and box loss. Additionally, the SHAP framework was utilized to gain a deeper understanding of the model's prediction criteria, making the decision-making process more transparent. This analysis provided significant insights into how the model considers various variables to detect diseases and pests.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.7
• /
• pp.271-278
• /
• 2021

As the current cyber attacks become more intelligent, the existing Intrusion Detection System is difficult for detecting intelligent attacks that deviate from the existing stored patterns. In an attempt to solve this, a model of a deep learning-based intrusion detection system that analyzes the pattern of intelligent attacks through data learning has emerged. Intrusion detection systems are divided into host-based and network-based depending on the installation location. Unlike network-based intrusion detection systems, host-based intrusion detection systems have the disadvantage of having to observe the inside and outside of the system as a whole. However, it has the advantage of being able to detect intrusions that cannot be detected by a network-based intrusion detection system. Therefore, in this study, we conducted a study on a host-based intrusion detection system. In order to evaluate and improve the performance of the host-based intrusion detection system model, we used the host-based Leipzig Intrusion Detection-Data Set (LID-DS) published in 2018. In the performance evaluation of the model using that data set, in order to confirm the similarity of each data and reconstructed to identify whether it is normal data or abnormal data, 1D vector data is converted to 3D image data. Also, the deep learning model has the drawback of having to re-learn every time a new cyber attack method is seen. In other words, it is not efficient because it takes a long time to learn a large amount of data. To solve this problem, this paper proposes the Siamese Convolutional Neural Network (Siamese-CNN) to use the Few-Shot Learning method that shows excellent performance by learning the little amount of data. Siamese-CNN determines whether the attacks are of the same type by the similarity score of each sample of cyber attacks converted into images. The accuracy was calculated using Few-Shot Learning technique, and the performance of Vanilla Convolutional Neural Network (Vanilla-CNN) and Siamese-CNN was compared to confirm the performance of Siamese-CNN. As a result of measuring Accuracy, Precision, Recall and F1-Score index, it was confirmed that the recall of the Siamese-CNN model proposed in this study was increased by about 6% from the Vanilla-CNN model.