• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.3
• /
• pp.121-126
• /
• 2020

Sound-based machine fault diagnosis is the automatic detection of abnormal sound in the acoustic emission signals of the machines. Conventional methods of using mathematical models were difficult to diagnose machine failure due to the complexity of the industry machinery system and the existence of nonlinear factors such as noises. Therefore, we want to solve the problem of machine fault diagnosis as a deep learning-based image classification problem. In the paper, we propose a CNN-based automatic machine fault diagnosis method using Spectrogram images. The proposed method uses STFT to effectively extract feature vectors from frequencies generated by machine defects, and the feature vectors detected by STFT were converted into spectrogram images and classified by CNN by machine status. The results show that the proposed method can be effectively used not only to detect defects but also to various automatic diagnosis system based on sound.

• The Korean Journal of Applied Statistics
• /
• v.35 no.1
• /
• pp.49-62
• /
• 2022

Realized volatility is well known to have long memory, strong association with other global financial markets and interdependences among macroeconomic indices such as exchange rate, oil price and interest rates. This paper proposes autoencoder factor-augmented heterogeneous autoregressive (AE-FAHAR) model for realized volatility forecasting. AE-FAHAR incorporates long memory using HAR structure, and exogenous variables into few factors summarized by autoencoder. Autoencoder requires intensive calculation due to its nonlinear structure, however, it is more suitable to summarize complex, possibly nonstationary high-dimensional time series. Our AE-FAHAR model is shown to have smaller out-of-sample forecasting error in empirical analysis. We also discuss pre-training, ensemble in autoencoder to reduce computational cost and estimation errors.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.349-354
• /
• 2022

In thermal power plants, boiler models have been used widely in evaluating logic configurations, performing system tuning and applying control theory, etc. Furthermore, proper plant models are needed to design the accurate controllers. Sometimes, mathematical models can not exactly describe a power plant due to time varying, nonlinearity, uncertainties and complexity of the thermal power plants. In this case, a neural network can be a useful method to estimate such systems. In this paper, the models of boiler steam system in a thermal power plant are developed by using a generalized regression neural network(GRNN). The models of the superheater, reheater, attemperator and drum are designed by using GRNN and the models are trained and validate with the real data obtained in 540[MW] power plant. The validation results showed that proposed models agree with actual outputs of the drum boiler well.

• Journal of Advanced Navigation Technology
• /
• v.26 no.3
• /
• pp.166-171
• /
• 2022

In this paper, we propose an algorithm to improve DoA(direction of arrival) estimation performance of the subspace-based method by generating high robustness correlation matrix of the signals incident on the uniformly linear array antenna. The existing subspace-based DoA estimation method estimates the DoA by obtaining a correlation matrix and dividing it into a signal subspace and a noise subspace. However, the component of the correlation matrix obtained from the low SNR and small number of snapshots inaccurately estimates the signal subspace due to the noise component of the antenna, thereby degrading the DoA estimation performance. Therefore a robust correlation matrix is generated by arranging virtual signal vectors obtained from the existing correlation matrix in a sliding manner. As a result of simulation using MUSIC and ESPRIT, which are representative subspace-based methods,, the computational complexity increased by less than 2.5% compared to the existing correlation matrix, but both MUSIC and ESPRIT based on RMSE 1° showed superior DoA estimation performance with an SNR of 3dB or more.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.4-4
• /
• 2023

C-band SAR 센서를 탑재한 수자원위성은 한반도 수자원 모니터링을 위해 개발되어 2025년 발사가 계획되어 있으며, 수변환경 및 부유물 탐지 및 다양한 활용이 기대되고 있다. 그 중 수변환경은 수변 생태계 안정성을 유지하는 역할을 담당하여 이에 대한 모니터링은 중요하다. s현장 관측 기반 탐지 방법과 비교하여 위성 원격탐사는 광범위한 지역을 반복적으로 관측하여, 연속적인 수변환경 및 부유물 정보를 제공할 수 있다. 이러한 특성에 기반하여 다양한 다중분광 및 SAR (Synthetic Aperture Radar) 위성 원격탐사 자료를 바탕으로 수변환경 및 부유물의 탐지 연구가 이루어졌다. 특히 단일 영상만을 사용하는 기법에 비해 다중분광 및 SAR 영상을 융합하여 높은 정확도를 보인 바 있다. 초기 연구에서는 임계값 알고리즘 또는 현장관측 기반의 부유물 농도와 위성 자료간의 선형관계를 분석하는 단순한 알고리즘이 주를 이루었으나, 최근에는 RF, CNN 등 보다 복잡하고 다양한 인공지능 알고리즘이 적용되어 높은 정확도로 해당 문제들을 해결하고 있다. 본 연구에서는 수자원위성 활용을 위해 인공지능 기반 수변환경 및 부유물 탐지 알고리즘을 개발하고자 한다. 수자원위성의 대체 자료로 유럽우주국의 Sentinel-1 A/B 위성의 C-band SAR 영상을 이용하였으며, 보조자료로 Sentinel-2 다중분광 영상을 이용하였다. 개발된 알고리즘은 수자원 관리를 위한 환경변화 탐지에 유용한 정보로 활용될 수 있을 것으로 기대된다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.299-299
• /
• 2023

토양수분(soil moisture)은 수문인자의 하나로서 토양 내에 함유된 물의 양을 의미하며, 그 총량은 미미하지만 대기와 지표면 사이에서 일어나는 복잡한 물순환과 에너지 교환을 이해하는데 있어 필수적이다. 현재 국내에서는 「수자원의 조사·계획 및 관리에 관한 법률」(이하 수자원법)에 근거해 토양수분량 관측이 이루어지고 있으며, 수자원 분야의 한국수자원조사기술원 외에도 농업, 임업 분야에서도 다양한 기관에서 지상관측소를 구축해 토양수분량을 측정하고 있다. 국내 지상관측소에서는 주로 지점규모(point scale)로 토양수분량을 관측하는 장비가 사용되고 있으며, 유전율식 장비인 TDR(Time Domain Reflectometry), FDR(Frequency Domain Reflectometry)이나 토양수분장력을 측정하는 장력계(Tensiometer)가 널리 쓰이고 있다. 수자원분야에서는 토양 내 수분의 양을 직관적으로 확인할 수 있는 유전율식 장비가 대중적으로 사용되고 있으며, 최근에는 우주선(Cosmic-Ray)으로부터 발생하는 고속중성자(Fast Neutron)를 통해 중규모 면단위(field scale) 토양수분량을 관측하는 장비인 CRNP(Cosmic-Ray Neutron Probe)에 대한 연구도 활발히 진행되고 있다. 이러한 장비는 주로 야외에 설치해 운영하고 있기 때문에 장비 훼손이나 전원공급의 어려움으로 결측이나 오측이 발생할 수 있다. 토양수분량 시계열자료의 결측이나 오측이 일반적인 감쇄기에 발생했다면 선형보간법으로도 간단히 보간할 수 있지만, 강우에 의한 상승기에 발생했다면 해당 강우사상에서의 토양수분량의 상한치를 알기 어려워 결측보간에 어려움이 있다. 본 연구에서는 토양수분량 시계열자료의 강우기간 결측을 보간하는 방법으로 누적분포함수 역변환 샘플링방법을 선택하였다. 연구에는 음성군(차곡리) 토양수분량 관측소 2021년 자료가 사용되었으며, 관측소 56개 지점 중 임의의 지점에 결측구간을 생성한 뒤 해당 지점과의 상관계수가 높은 지점의 누적분포함수를 이용해 역변환 샘플링 방식으로 임의 지점의 결측을 보간하고 그 결과를 기존값과 비교해 보간 방법의 정확도를 평가하였다.

• Journal of Creative Information Culture
• /
• v.5 no.3
• /
• pp.345-353
• /
• 2019

Several studies have been proposed to improve storage space efficiency by expressing binary trie data structure as a linear binary bit string. Compact binary tree approach generated using one binary trie increases the key search time significantly as the binary bit string becomes very long as the size of the input key set increases. In order to reduce the key search range, a hierarchical compact binary tree technique that hierarchically expresses several small binary compact trees has been proposed. The search time increases proportionally with the number and length of binary bit streams. In this paper, we generate several binary compact trees represented by full binary tries hierarchically. The search performance is improved by allowing a path for the binary bit string corresponding to the search range to be determined through simple numeric conversion. Through the performance evaluation using the worst time and space complexity calculation, the proposed method showed the highest performance for retrieval and key insertion or deletion. In terms of space usage, the proposed method requires about 67% ~ 68% of space compared to the existing methods, showing the best space efficiency.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.4
• /
• pp.47-58
• /
• 2010

Seepage rate through the core zone of rockfill dam, estimated from graphical technique and the equation by Sakamoto (1998), is different from the real condition because of neglecting unsaturated flow. With existing method to estimate total seepage rate, it is difficult to understand the tendency of total seepage rate changes by reservoir water level change. Steady state seepage rate and the factors affecting the time needed to attain to changes of reservoir water level and saturated hydraulic conductivity and unsaturated hydraulic properties of core material are analysed thorough the 2-D steady and unsteady state seepage analyses of Soyanggang dam. Numerical results revealed that the seepage rate can be expressed by the linear equation form and the value of unsaturated soil parameter n is the most important factor affecting the seepage rate and the time needed to attain steady state. The estimation method presented in this study can be used by the designer and the personnel of dam safety for convenient estimation of seepage rate and quantitative analysis of measured seepage rate without 2-D and 3-D numerical analyses.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.1
• /
• pp.135-143
• /
• 2009

Recently, in the mobile device, the increase of the need for encoding and decoding of high-resolution images requires an efficient implementation of the image codec. This paper proposes a time-optimized color conversion method for the JPEG decoder, which reduces the number of calculations in the color conversion by the rearrangement of arithmetic operations being possible due to the linearity of the IDCT and the color conversion matrices and brings down the time complexity of the color conversion itself by the integer mapping replacing floating-point operations to the optimal fixed-point shift and addition operations, eventually reducing the time complexity of the JPEG decoder. And the proposed method compensates a decline of image quality incurred by the quantification error of the operation arrangement and the integer mapping by using the multi-mode chrominance reconstruction. The performance evaluation performed on the development platform of embedded systems showed that, compared to previous color conversion methods, the proposed method greatly reduces the image decoding time, minimizing the distortion of decoded images.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.31-38
• /
• 2009

Lens distortion is inevitable phenomenon in machine vision system. More and more distortion phenomenon is occurring in order to choice of lens for minimizing cost and system size. As shown above, correction of lens distortion is critical issue. However previous lens correction methods using camera model have problem such as nonlinear property and complicated operation. And recent lens correction methods using neural network also have accuracy and efficiency problem. In this study, I propose new algorithms for correction of lens distortion. Distorted image is divided based on the distortion quantity using k-means. And each divided image region is corrected by using neural network. As a result, the proposed algorithms have better accuracy than previous methods without image division.