• Nuclear Engineering and Technology
• /
• v.55 no.4
• /
• pp.1507-1517
• /
• 2023

Centrifugal pumps are a crucial part of nuclear power plants, and their dependable and safe operation is crucial to the security of the entire facility. Cavitation will cause the centrifugal pump to violently vibration with the large number of vacuoles generated, which not only affect the hydraulic performance of the centrifugal pump but also cause structural damage to the impeller, seriously affecting the operational safety of nuclear power plants. A closed cavitation test bench of a centrifugal pump is constructed, and a method for precisely identifying the cavitation state is proposed based on Complementary Ensemble Empirical Mode Decomposition (CEEMD) and Deep Residual Shrinkage Network (DRSN). First, we compared the cavitation sensitivity of pressure fluctuation, vibration, and liquid-borne noise and decomposed the liquid-borne noise by CEEMD to capture cavitation characteristics. The decomposition results are sent into a 12-layer deep residual shrinkage network (DRSN) for cavitation identification training. The results demonstrate that the liquid-borne noise signal is the most cavitation-sensitive signal, and the accuracy of CEEMD-DRSN to identify cavitation at different stages of centrifugal pumps arrives at 94.61%

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.153-153
• /
• 2020

우리나라에선 크고 작은 가뭄 피해가 자주 일어나고 있으며 최근엔 유래 없는 다년가뭄이 발생하면서 가뭄에 대한 경각심이 커지고 있다. 가뭄에 적절하게 대응하여 피해를 경감시키기 위해서는 신뢰도 높은 가뭄 예측이 선행되어야 한다. 이에 본 연구는 앙상블 예측과 베이즈이론(Bayes' theorem)을 수문학적 가뭄지수 중 하나인 SRI(Standardized Runoff Index)에 적용해 가뭄 확률 전망을 실시했으며 이를 EDP(Ensemble Drought Prediction)라고 칭하였다. 국내 8개 댐유역에서 EDP를 생성하고 개선하는 과정은 다음과 같이 진행된다. 우선 TANK모형을 활용한 1개월 선행 유량 예측(Ensemble Streamflow Prediction, ESP)의 결과를 SRI로 변환하여 EDP 확률분포를 생성한다. 그런 다음, EDP를 개선하기 위해 그 기초인 ESP에서 미흡한 토양수분 초기조건을 보완하고자 베이즈이론을 활용했다. APCC(APEC Climate Center)의 위성 관측 SMI(Soil Moisture Index) 자료로 SRI와의 회귀식을 구축, 이를 우도함수로 정의해 사전 EDP 분포를 업데이트한 EDP+ 확률분포를 생성했다. 그 결과, EDP와 EDP+ 모두 심도가 깊은 가뭄을 전망할수록 예측력이 기후학적 예측보다 좋지 않았다. 그럼에도 우도함수로 사용한 회귀식의 정확도가 높을수록 EDP+의 정확도도 향상되는 경향이 나타났으며, 이는 베이즈이론을 사용한다면 가뭄 확률 전망을 개선할 수 있다는 것을 의미하고 있다. 하지만, 확정 전망 정확도는 확률 전망 정확도와는 관계가 없었는데 이는 확정 전망과 확률 전망이 본질적으로 다르기 때문인 것으로 사료된다.

• International Journal of Computer Science & Network Security
• /
• v.24 no.5
• /
• pp.64-72
• /
• 2024

In recent times cyber attackers can use Artificial Intelligence (AI) to boost the sophistication and scope of attacks. On the defense side, AI is used to enhance defense plans, to boost the robustness, flexibility, and efficiency of defense systems, which means adapting to environmental changes to reduce impacts. With increased developments in the field of information and communication technologies, various exploits occur as a danger sign to cyber security and these exploitations are changing rapidly. Cyber criminals use new, sophisticated tactics to boost their attack speed and size. Consequently, there is a need for more flexible, adaptable and strong cyber defense systems that can identify a wide range of threats in real-time. In recent years, the adoption of AI approaches has increased and maintained a vital role in the detection and prevention of cyber threats. In this paper, an Ensemble Deep Restricted Boltzmann Machine (EDRBM) is developed for the classification of cybersecurity threats in case of a large-scale network environment. The EDRBM acts as a classification model that enables the classification of malicious flowsets from the largescale network. The simulation is conducted to test the efficacy of the proposed EDRBM under various malware attacks. The simulation results show that the proposed method achieves higher classification rate in classifying the malware in the flowsets i.e., malicious flowsets than other methods.

• Journal of Information Processing Systems
• /
• v.20 no.4
• /
• pp.501-513
• /
• 2024

In recent years, graph neural networks (GNNs) have been extensively used to analyze graph data across various domains because of their powerful capabilities in learning complex graph-structured data. However, recent research has focused on improving the performance of a single GNN with only two or three layers. This is because stacking layers deeply causes the over-smoothing problem of GNNs, which degrades the performance of GNNs significantly. On the other hand, ensemble methods combine individual weak models to obtain better generalization performance. Among them, gradient boosting is a powerful supervised learning algorithm that adds new weak models in the direction of reducing the errors of the previously created weak models. After repeating this process, gradient boosting combines the weak models to produce a strong model with better performance. Until now, most studies on GNNs have focused on improving the performance of a single GNN. In contrast, improving the performance of GNNs using multiple GNNs has not been studied much yet. In this paper, we propose gradient boosted graph neural networks (GBGNN) that combine multiple shallow GNNs with gradient boosting. We use shallow GNNs as weak models and create new weak models using the proposed gradient boosting-based loss function. Our empirical evaluations on three real-world datasets demonstrate that GBGNN performs much better than a single GNN. Specifically, in our experiments using graph convolutional network (GCN) and graph attention network (GAT) as weak models on the Cora dataset, GBGNN achieves performance improvements of 12.3%p and 6.1%p in node classification accuracy compared to a single GCN and a single GAT, respectively.

• KIISE Transactions on Computing Practices
• /
• v.21 no.1
• /
• pp.64-69
• /
• 2015

Place awareness is an essential for location-based services that are widely provided to smartphone users. However, traditional GPS-based methods are only valid outdoors where the GPS signal is strong and also require symbolic place information of the physical location. In this paper, environmental sounds and images are used to recognize important aspects of each place. The proposed method extracts feature vectors from visual, auditory and location data recorded by a smartphone with built-in camera, microphone and GPS sensors modules. The heterogeneous feature vectors were then learned by an ensemble learning method that learns each group of feature vectors for each classifier respectively and votes to produce the highest weighted result. The proposed method is evaluated for place recognition using a data group of 3000 samples in six places and the experimental results show a remarkably improved recognition accuracy when using all kinds of sensory data comparing to results using data from a single sensor or audio-visual integrated data only.

• Journal of the Korean Chemical Society
• /
• v.53 no.6
• /
• pp.653-662
• /
• 2009

Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

• Journal of Korean Society for Geospatial Information Science
• /
• v.22 no.4
• /
• pp.47-52
• /
• 2014

Species distribution models have been widely applied in order to assess biodiversity, design reserve, manage habitat and predict climate change. However, SDMs has been used restrictively to the public and policy sectors owing to model uncertainty. Recent studies on ensemble and consensus models have been increased to reduce model uncertainty. This paper was carried out single model and multi model for Corylopsis coreana and compares two models. First, model evaluation was used AUC, kappa and TSS. TSS was the most effective method because it was easy to compare several models and convert binary maps. Second, both single and ensemble model show good performance and RF, Maxent and GBM was evaluated higher, GAM and SRE was evaluated lower relatively. Third, ensemble model tended to overestimate over single model. This problem can be solved by the suitable model selection and weighting through collaboration between field experts and modeler. Finally, we should identify causes and magnitude of model uncertainty and improve data quality and model methods in order to apply special decision-making support system and conservation planning, and when we make policy decisions using SDMs, we should recognize uncertainty and risk.

• Geophysics and Geophysical Exploration
• /
• v.26 no.4
• /
• pp.211-228
• /
• 2023

Ground-penetrating radar (GPR) surveys are commonly used to monitor embankments, which is a nondestructive geophysical method. The results of GPR surveys can be complex, depending on the situation, and data processing and interpretation are subject to expert experiences, potentially resulting in false detection. Additionally, this process is time-intensive. Consequently, various studies have been undertaken to detect cavities in GPR survey data using deep learning methods. Deep-learning-based approaches require abundant data for training, but GPR field survey data are often scarce due to cost and other factors constaining field studies. Therefore, in this study, a deep- learning-based model was developed for embankment GPR survey cavity detection using data augmentation strategies. A dataset was constructed by collecting survey data over several years from the same embankment. A you look only once (YOLO) model, commonly used in computer vision for object detection, was employed for this purpose. By comparing and analyzing various strategies, the optimal data augmentation approach was determined. After initial model development, a stepwise process was employed, including box clustering, transfer learning, self-ensemble, and model ensemble techniques, to enhance the final model performance. The model performance was evaluated, with the results demonstrating its effectiveness in detecting cavities in embankment GPR survey data.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.79-89
• /
• 2016

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.

• The Korean Journal of Applied Statistics
• /
• v.32 no.5
• /
• pp.671-681
• /
• 2019

Studies on predicting arrhythmia using machine learning have been actively conducted with increasing number of arrhythmia patients. Existing studies have predicted arrhythmia based on multivariate data of feature variables extracted from RR interval data at a specific time point. In this study, we consider that the pattern of the heart state changes with time can be important information for the arrhythmia prediction. Therefore, we investigate the usefulness of predicting the arrhythmia with multivariate time series data obtained by extracting and accumulating the multivariate vectors of the feature variables at various time points. When considering 1-nearest neighbor classification method and its ensemble for comparison, it is confirmed that the multivariate time series data based method can have better classification performance than the multivariate data based method if we select an appropriate time series distance function.