• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.292-292
• /
• 2021

최근 데이터 과학의 획기적인 발전으로 딥러닝(Deep Learning) 알고리즘이 개발되어 다양한 분야에 널리 적용되고 있다. 본 연구에서는 인공신경망 중 하나인 LSTM(Long-Short Term Memory) 네트워크를 기반으로 정확한 댐유입량 예측을 수행하는 SWLSTM 모델을 제안하였다. SWLSM은 모델의 정확도를 개선하기 위해 세 가지 주요 아이디어를 채택하였다. (1) 통계적 속성 (PACF) 및 교차 상관 함수(CCF)를 사용하여 적절한 입력 변수와 시퀀스 길이를 결정하였다. (2) 선택된 입력 예측 변수 시계열을 웨이블릿 변환(WT)을 사용하여 하위 시계열로 분해한다. (3) k-folds cross validation 및 random search 기법을 사용하여 LSTM의 하이퍼 매개변수들을 효율적으로 최적화하고 검증한다. 제안된 SWLSTM의 효과는 한강 유역 5개 댐의 시단위/일단위/월단위 유입량을 예측하고 과거 자료와 비교함으로써 검증하였다. 모델의 정확도는 다양한 평가 메트릭(R2, NSE, MAE, PE)이 사용하였으며, SWLSTM은 모든 경우에서 LSTM 모델을 능가하였다. (평가 지표는 약 30 ~ 80 % 더 나은 성능을 보여줌). 본 연구의 결과로부터, 올바른 입력 변수와 시퀀스 길이의 선택이 모델 학습의 효율성을 높이고 노이즈를 줄이는 데 효과적임을 확인하였다. WT는 홍수 첨두와 같은 극단적인 값을 예측하는 데 도움이 된다. k-folds cross validation 및 random search 기법을 사용하면 모델의 하이퍼 매개변수를 효율적으로 설정할 수 있다. 본 연구로부터 댐 유입량을 정확하게 예측한다면 정책 입안자와 운영자가 저수지 운영, 계획 및 관리에 도움이 될 것이다.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.617-623
• /
• 2024

Identifying emotions through speech poses a significant challenge due to the complex relationship between language and emotions. Our paper aims to take on this challenge by employing feature engineering to identify emotions in speech through a multimodal classification task involving both speech and text data. We evaluated two classifiers-Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM)-both integrated with a BERT-based pre-trained model. Our assessment covers various performance metrics (accuracy, F-score, precision, and recall) across different experimental setups). The findings highlight the impressive proficiency of two models in accurately discerning emotions from both text and speech data.

• Journal of Korean Society of Disaster and Security
• /
• v.15 no.3
• /
• pp.89-99
• /
• 2022

Bridges are a key infrastructure that underpins economic and social activities. In Korea, bridges began to be built with economic development in the 1970s and were built intensively in the 1980s and 1990s. In recent years, as the number of bridges with a service life of more than 30 years is increasing, continuous maintenance is required to ensure the safety of the bridges. In particular, in order to cope with the aging of bridges, research on technology development such as maintenance using ICT technology, preventive maintenance, life cycle cost reduction, and long life bridge is being actively promoted. This paper presents the results of correlation analysis based on the safety evaluation data of bridges as part of the research on the development of a model for estimating load-carrying capacity of bridges. As a analysis result, indicators highly correlated with the load-carrying capacity of the bridge was derived.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.6
• /
• pp.646-661
• /
• 2003

It is not easy that we visualize the large volume data stored in the every client computers of the web environment. One solution is as follows. First we compress volume data, second store that in the database server, third transfer that to client computer, fourth visualize that with direct-volume-rendering in the client computer. In this case, we usually use wavelet transform for compressing large data. This paper reports the experiments for acquiring the wavelet bases and the compression ratios fit for the above processing paradigm. In this experiments, we compress the volume data Engine, CThead, Bentum into 50%, 10%, 5%, 1%, 0.1%, 0.03% of the total data respectively using Harr, Daubechies4, Daubechies12 and Daubechies20 wavelets, then visualize that with direct-volume-rendering, afterwards evaluate the images with eyes and image comparison metrics. When compression ratio being low the performance of Harr wavelet is better than the performance of the other wavelets, when compression ratio being high the performance of Daubechies4 and Daubechies12 is better than the performance of the other wavelets. When measuring with eyes the good compression ratio is about 1% of all the data, when measuring with image comparison metrics, the good compression ratio is about 5-10% of all the data.

• Phonetics and Speech Sciences
• /
• v.16 no.1
• /
• pp.67-76
• /
• 2024

Recent advances in text-to-speech (TTS) technology have significantly improved the quality of synthesized speech, reaching a level where it can closely imitate natural human speech. Especially, TTS models offering various voice characteristics and personalized speech, are widely utilized in fields such as artificial intelligence (AI) tutors, advertising, and video dubbing. Accordingly, in this paper, we propose a one-shot multi-speaker TTS system that can ensure acoustic diversity and synthesize personalized voice by generating speech using unseen target speakers' utterances. The proposed model integrates a speaker encoder into a TTS model consisting of the FastSpeech2 acoustic model and the HiFi-GAN vocoder. The speaker encoder, based on the pre-trained RawNet3, extracts speaker-specific voice features. Furthermore, the proposed approach not only includes an English one-shot multi-speaker TTS but also introduces a Korean one-shot multi-speaker TTS. We evaluate naturalness and speaker similarity of the generated speech using objective and subjective metrics. In the subjective evaluation, the proposed Korean one-shot multi-speaker TTS obtained naturalness mean opinion score (NMOS) of 3.36 and similarity MOS (SMOS) of 3.16. The objective evaluation of the proposed English and Korean one-shot multi-speaker TTS showed a prediction MOS (P-MOS) of 2.54 and 3.74, respectively. These results indicate that the performance of our proposed model is improved over the baseline models in terms of both naturalness and speaker similarity.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.565-571
• /
• 2023

With the introduction of the post-verification system, the measurement of floor impact noise performance on-site has become mandatory, and the evaluation method has changed. To track the performance changes since the policy implementation, research is needed on how the characteristics of heavyweight impact sound change according to the varied evaluation method. In this study, we analyzed the contribution rate of the frequency band-specific sound pressure level on the single-number quantity for a multi-family housing unit with the same floor plan and floor structure, comprising 59 households, based on the changed impact sources and evaluation indicators. It is difficult to compare simply because the method of calculating contributions by frequency band according to the single-day evaluation is different, but the average contribution rate of 63 Hz was 80.8 % in the evaluation method before the introduction of the post-confirmation system (Tire measurement and evaluated as L'_i,Fmax,AW), and the average contribution rate of 125 Hz was 19.2 %. The current evaluation method (rubber ball measurement and evaluation as L'_iA,Fmax) shows that the contribution rate has decreased to 33.1 % on average at 50 Hz ~ 80 Hz, 58.7 % on average at 100 Hz ~ 160 Hz, 6.9 % on average at 200 Hz ~ 315 Hz, and 1.3 % on average at 400 Hz ~ 630 Hz. This result is a case analysis for the target apartment house, and it is necessary to analyze measurement data for more diverse apartment houses.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.4
• /
• pp.35-42
• /
• 2013

In this paper, the analog performance of FinFET structure was estimated by extracting the DC/AC characteristics of the 22 nm process FinFET structures with different layout considering spacer and SEG using 3D device simulator, Sentaurus. Based on the analysis results, layout methods to enhance the analog performance of multi-fin FinFET structures are proposed. By adding the spacer and SEG structures, the drive current of 1-fin FinFET increases. However, the unity gain frequency, $f_T$, reduces by 19.4 % due to the increase in the total capacitance caused by the added spacer. If the process element is not included in multi-fin FinFET, replacing 1-finger with 2-finger structure brings approximately 10 % of analog performance improvement. Considering the process factors, we propose methods to maximize the analog performance by optimizing the interconnect and gate structures.

• The Journal of the Korea Contents Association
• /
• v.20 no.6
• /
• pp.556-578
• /
• 2020

There are many different researches about how to analyze issues based on real-time news streams. But, there are few researches which analyze issues hierarchically from news articles and even a previous research of hierarchical issue analysis make clustering speed slower as the increment of news articles. In this paper, we propose a hierarchical and incremental clustering for semi real-time issue analysis on news articles. We trained siamese neural network based weighted cosine similarity model, applied this model to k-means algorithm which is used to make word clusters and converted news articles to document vectors by using these word clusters. Finally, we initialized an issue cluster tree from document vectors, updated this tree whenever news articles happen, and analyzed issues in semi real-time. Through the experiment and evaluation, we showed that up to about 0.26 performance has been improved in terms of NMI. Also, in terms of speed of incremental clustering, we also showed about 10 times faster than before.

• 한국방재학회:학술대회논문집
• /
• 2011.02a
• /
• pp.76-76
• /
• 2011

현재까지 시공된 사장교 중, 주경간이 가장 긴 교량은 중국의 수통대교(1088m)이다. 이에 버금가는 사장교로 홍콩의 스톤커터교(1018m) 역시 주경간장이 1000m가 넘는다. 바야흐로 사장교 역시 주경간 1000m의 시대가 열린 것이다. 우리나라 역시 세계적 흐름에 맞추어 주경간 800m의 인천대교(세계 5위)를 시공한바 있다. 이와 같이 교량의 초장대화는, 교량 건설 분야에서 기술경쟁력의 지표가 될 뿐만 아니라 세계 건설 시장의 큰 흐름이라고 할 수 있다. 이에 본 연구는 세계적 추세에 발맞추어, 국내 각계의 건설 전문가들이 모여 만든 초장대 교량 사업단의 기술 혁신 사업의 일환으로 이루어졌다. 교량이 장대화 되면서 바람의 의한 영향이 중요해진다는 것은 주지의 사실이다. 특히 사장교와 현수교 같은 특수 교량의 경우, 정적 및 동적 내풍 성능이 반드시 고려되어야만 한다. 본 연구에서는 주경간 1200m의 사장교를 가정하고, 이 사장교의 내풍 단면을 개발, 그 단면에 대한 정적 및 동적 내풍 성능을 평가하고자 하였다. 정적 내풍 성능으로는 단면의 형상에 따른 풍하중을 파악하고자 했으며, 동적 내풍 성능으로는 풍속에 따른 교량의 연직방향 변위 및 플러터 속도를 파악하고자 하였다. 이 실험은 추후에 3차원 전교모형실험의 기본 데이터로 활용하였다. 본 실험을 통해 개발된 단면의 등류 및 난류 상태에서의 영각별 정적 공기력계수를 계산해내었고, 설계풍속이 54.7m/s일때 한계풍속 65.64m/s(거마대교 기준)하에서의 중앙 경간의 풍속별 평균 변위를 측정하였으며, 이를 토대로 이 교량의 영각별 플러터 속도를 계산해 내었다.

• Journal of Navigation and Port Research
• /
• v.32 no.6
• /
• pp.439-445
• /
• 2008

Various turning tests were carried out according to the rudder angle, turning direction, and the speed etc. with the ship's maneuverability measuring system on the training ship HANBADA. After that they were compared with each other on the turning circle, maneuvering performance index and the distance of new course, and then found out that they were satisfied with the IMO maneuvering standards. And the turning circles of port were smaller than those of starboard with all the rudder angles and maneuvering indexes such as K and T were relatively bigger than other vessels. Also, the distance cf new course was measured to $125{\sim}300m$ in case of the new course on $30^{\circ}{\sim}90^{\circ}$. All of these results will be helpful to escape from collision and to alter course on coastal voyage.