• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• Nuclear Engineering and Technology
• /
• v.54 no.5
• /
• pp.1825-1834
• /
• 2022

Performing high-fidelity computational fluid dynamics (HF-CFD) to predict the flow and heat transfer state of the coolant in the reactor core is expensive, especially in scenarios that require extensive parameter search, such as uncertainty analysis and design optimization. This work investigated the performance of utilizing a multi-fidelity reduced-order model (MF-ROM) in PWR rod bundles simulation. Firstly, basis vectors and basis vector coefficients of high-fidelity and low-fidelity CFD results are extracted separately by the proper orthogonal decomposition (POD) approach. Secondly, a surrogate model is trained to map the relationship between the extracted coefficients from different fidelity results. In the prediction stage, the coefficients of the low-fidelity data under the new operating conditions are extracted by using the obtained POD basis vectors. Then, the trained surrogate model uses the low-fidelity coefficients to regress the high-fidelity coefficients. The predicted high-fidelity data is reconstructed from the product of extracted basis vectors and the regression coefficients. The effectiveness of the MF-ROM is evaluated on a flow and heat transfer problem in PWR fuel rod bundles. Two data-driven algorithms, the Kriging and artificial neural network (ANN), are trained as surrogate models for the MF-ROM to reconstruct the complex flow and heat transfer field downstream of the mixing vanes. The results show good agreements between the data reconstructed with the trained MF-ROM and the high-fidelity CFD simulation result, while the former only requires to taken the computational burden of low-fidelity simulation. The results also show that the performance of the ANN model is slightly better than the Kriging model when using a high number of POD basis vectors for regression. Moreover, the result presented in this paper demonstrates the suitability of the proposed MF-ROM for high-fidelity fixed value initialization to accelerate complex simulation.

• Journal of Broadcast Engineering
• /
• v.28 no.1
• /
• pp.132-140
• /
• 2023

In this paper, we propose a method for performance enhancement of speech declipping using clipping detector. Clipping occurs when the input speech level exceeds the dynamic range of microphone, and it significantly degrades the speech quality. Recently, many methods for high-performance speech declipping based on machine learning have been developed. However, they often deteriorate the speech signal because of degradation in signal reconstruction process when the degree of clipping is not high. To solve this problem, we propose a new approach that combines the declipping network and clipping detector, which enables a selective declipping operation depending on the clipping level and provides high-quality speech in all clipping levels. We measured the declipping performance using various metrics and confirmed that the proposed method improves the average performance over all clipping levels, compared with the conventional methods, and greatly improves the performance when the clipping distortion is small.

• Imaging Science in Dentistry
• /
• v.52 no.4
• /
• pp.351-357
• /
• 2022

Purpose: Convolutional neural networks (CNNs) have rapidly emerged as one of the most promising artificial intelligence methods in the field of medical and dental research. CNNs can provide an effective diagnostic methodology allowing for the detection of early-staged diseases. Therefore, this study aimed to evaluate the performance of a deep CNN algorithm for apical lesion segmentation from panoramic radiographs. Materials and Methods: A total of 1000 panoramic images showing apical lesions were separated into training (n=800, 80%), validation (n=100, 10%), and test (n=100, 10%) datasets. The performance of identifying apical lesions was evaluated by calculating the precision, recall, and F1-score. Results: In the test group of 180 apical lesions, 147 lesions were segmented from panoramic radiographs with an intersection over union (IoU) threshold of 0.3. The F1-score values, as a measure of performance, were 0.828, 0.815, and 0.742, respectively, with IoU thresholds of 0.3, 0.4, and 0.5. Conclusion: This study showed the potential utility of a deep learning-guided approach for the segmentation of apical lesions. The deep CNN algorithm using U-Net demonstrated considerably high performance in detecting apical lesions.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.2-2
• /
• 2023

기후변화가 가속화로 인해 수재해의 빈도와 강도 예측이 어려워짐에 따라 실시간 홍수 모니터링에 대한 수요가 증가하고 있다. 합성개구레이다는 광원과 날씨에 무관하게 촬영이 가능하여 수재해 발생시에도 영상을 확보할 수 있다. 합성개구레이다를 활용한 수체 탐지 알고리즘 개발이 활발히 연구되어 왔고, 딥러닝의 발달로 CNN을 활용하여 높은 정확도로 수체 탐지가 기능해졌다. 하지만, CNN 기반 수체 탐지 모델은 훈련시 높은 정량적 정확성 지표를 달성하여도 추론 후 정성적 평가시 경계와 소하천에 대한 탐지 정확성이 떨어진다. 홍수 모니터링에서 특히 중요한 정보인 경계와 좁은 하천에 대해서 정확성이 떨어짐에 따라 실생활 적용이 어렵다. 이에 경계를 강화한 적대적 학습 기반의 수체 탐지 모델을 개발하여 더 세밀하고 정확하게 탐지하고자 한다. 적대적 학습은 생성적 적대 신경망(GAN)의 두 개의 모델인 생성자와 판별자가 서로 관여하며 더 높은 정확도를 달성할 수 있도록 학습이다. 이러한 적대적 학습 개념을 수체 탐지 모델에 처음으로 도입하여, 생성자는 실제 라벨 데이터와 유사하게 수체 경계와 소하천까지 탐지하고자 학습한다. 반면 판별자는 경계 거리 변환 맵과 합성개구레이다 영상을 기반으로 라벨데이터와 수체 탐지 결과를 구분한다. 경계가 강화될 수 있도록, 면적과 경계를 모두 고려할 수 있는 손실함수 조합을 구성하였다. 제안 모델이 경계와 소하천을 정확히 탐지하는지 판단하기 위해, 정량적 지표로 F1-score를 사용하였으며, 육안 판독을 통해 정성적 평가도 진행하였다. 기존 U-Net 모델이 탐지하지 못하던 영역에 대해 제안한 경계 강화 적대적 수체 탐지 모델이 수체의 세밀한 부분까지 탐지할 수 있음을 증명하였다.

• Nuclear Engineering and Technology
• /
• v.55 no.8
• /
• pp.3030-3038
• /
• 2023

Nowadays, mobile robots have been used to search for uncontrolled radioactive source in indoor environments to avoid radiation exposure for technicians. However, in the indoor environments, especially in the presence of obstacles, how to make the robots with limited sensing capabilities automatically search for the radioactive source remains a major challenge. Also, the source search efficiency of robots needs to be further improved to meet practical scenarios such as limited exploration time. This paper proposes an automatic source search strategy, abbreviated as ACA: the location of source is estimated by a convolutional neural network (CNN), and the path is planned by the A-star algorithm. First, the search area is represented as an occupancy grid map. Then, the radiation dose distribution of the radioactive source in the occupancy grid map is obtained by Monte Carlo (MC) method simulation, and multiple sets of radiation data are collected through the eight neighborhood self-avoiding random walk (ENSAW) algorithm as the radiation data set. Further, the radiation data set is fed into the designed CNN architecture to train the network model in advance. When the searcher enters the search area where the radioactive source exists, the location of source is estimated by the network model and the search path is planned by the A-star algorithm, and this process is iterated continuously until the searcher reaches the location of radioactive source. The experimental results show that the average number of radiometric measurements and the average number of moving steps of the ACA algorithm are only 2.1% and 33.2% of those of the gradient search (GS) algorithm in the indoor environment without obstacles. In the indoor environment shielded by concrete walls, the GS algorithm fails to search for the source, while the ACA algorithm successfully searches for the source with fewer moving steps and sparse radiometric data.

• Nuclear Engineering and Technology
• /
• v.56 no.1
• /
• pp.160-166
• /
• 2024

The inversion of radiation field distribution is of great significance in the decommissioning sites of nuclear facilities. However, the radiation fields often contain multiple mixtures of radionuclides, making the inversion extremely difficult and posing a huge challenge. Many radiation field reconstruction methods, such as Kriging algorithm and neural network, can not solve this problem perfectly. To address this issue, this paper proposes an optimized inverse distance weighted (IDW) interpolation algorithm for reconstructing the gamma radiation field. The algorithm corrects the difference between the experimental and simulated scenarios, and the data is preprocessed with normalization to improve accuracy. The experiment involves setting up gamma radiation fields of three Co-60 radioactive sources and verifying them by using the optimized IDW algorithm. The results show that the mean absolute percentage error (MAPE) of the reconstruction result obtained by using the optimized IDW algorithm is 16.0%, which is significantly better than the results obtained by using the Kriging method. Importantly, the optimized IDW algorithm is suitable for radiation scenarios with multiple radioactive sources, providing an effective method for obtaining radiation field distribution in nuclear facility decommissioning engineering.

• Nuclear Engineering and Technology
• /
• v.56 no.6
• /
• pp.2343-2351
• /
• 2024

Accurately predicting the thermal hydraulic parameters of a transient reactor core under different working conditions is the first step toward reactor safety. Mass flow rate and temperature are important parameters of core thermal hydraulics, which have often been modeled as time series prediction problems. This study aims to achieve accurate and continuous prediction of core thermal hydraulic parameters under instantaneous conditions, as well as test the feasibility of a newly constructed gated recurrent unit (GRU) model based on the soft attention mechanism for core parameter predictions. Herein, the China Experimental Fast Reactor (CEFR) is used as the research object, and CEFR 1/2 core was taken as subject to carry out continuous predictive analysis of thermal parameters under transient conditions., while the subchannel analysis code named SUBCHANFLOW is used to generate the time series of core thermal-hydraulic parameters. The GRU model is used to predict the mass flow and temperature time series of the core. The results show that compared to the adaptive radial basis function neural network, the GRU network model produces better prediction results. The average relative error for temperature is less than 0.5 % when the step size is 3, and the prediction effect is better within 15 s. The average relative error of mass flow rate is less than 5 % when the step size is 10, and the prediction effect is better in the subsequent 12 s. The GRU model not only shows a higher prediction accuracy, but also captures the trends of the dynamic time series, which is useful for maintaining reactor safety and preventing nuclear power plant accidents. Furthermore, it can provide long-term continuous predictions under transient reactor conditions, which is useful for engineering applications and improving reactor safety.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.2028-2042
• /
• 1994

Properly defining signal and noise in a self-organizing system like ART(Adaptive Resonance Theory) neural network model raises a number of subtle issues. Pattern context must enter the definition so that input features, treated as irrelevant noise when they are embedded in a given input pattern, may be treated as informative signals when they are embedded in a different input pattern. The ATR automatically self-scales their computational units to embody context and learning dependent definitions of a signal and noise and there is no problem in categorizing input pattern that have features similar in nature. However, when we have imput patterns that have features that are different in size and nature, the use of only one vigilance parameter is not enough to differentiate a signal from noise for a good categorization. For example, if the value fo vigilance parameter is large, then noise may be processed as an informative signal and unnecessary categories are generated: and if the value of vigilance parameter is small, an informative signal may be ignored and treated as noise. Hence it is no easy to achieve a good pattern categorization. To overcome such problems, a Coupled-ART neural network capable of modularized categorization of patterns is proposed. The Coupled-ART has two layer of tightly coupled modules. the upper and the lower. The lower layer processes the global features of a pattern and the structural features, separately in parallel. The upper layer combines the categorized outputs from the lower layer and categorizes the combined output, Hence, due to the modularized categorization of patterns, the Coupled-ART classifies patterns more efficiently than the ART1 model.