• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1610-1629
• /
• 2021

Failures frequently occurred in manufacturing machines due to complex and changeable manufacturing environments, increasing the downtime and maintenance costs. This manuscript develops a novel deep learning-based method named Multi-Domain Convolutional Neural Network (MDCNN) to deal with this challenging task with vibration signals. The proposed MDCNN consists of time-domain, frequency-domain, and statistical-domain feature channels. The Time-domain channel is to model the hidden patterns of signals in the time domain. The frequency-domain channel uses Discrete Wavelet Transformation (DWT) to obtain the rich feature representations of signals in the frequency domain. The statistic-domain channel contains six statistical variables, which is to reflect the signals' macro statistical-domain features, respectively. Firstly, in the proposed MDCNN, time-domain and frequency-domain channels are processed by CNN individually with various filters. Secondly, the CNN extracted features from time, and frequency domains are merged as time-frequency features. Lastly, time-frequency domain features are fused with six statistical variables as the comprehensive features for identifying the fault. Thereby, the proposed method could make full use of those three domain-features for fault diagnosis while keeping high distinguishability due to CNN's utilization. The authors designed massive experiments with 10-folder cross-validation technology to validate the proposed method's effectiveness on the CWRU bearing data set. The experimental results are calculated by ten-time averaged accuracy. They have confirmed that the proposed MDCNN could intelligently, accurately, and timely detect the fault under the complex manufacturing environments, whose accuracy is nearly 100%.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(4)
• /
• pp.405-408
• /
• 2002

LSP(Line Spectrum Pairs) Parameter is used for speech analysis in vocoders or recognizers since it has advantages of constant spectrum sensitivity. low spectrum distortion and easy linear interpolation. However the method of transforming LPC(Linear Predictive Coding) into LSP is so complex that it takes much time to compute. Among conventional methods, the real root method is considerably simpler than others, but nevertheless, it still suffers from its jndeterministic computation time because the root searching is processed sequentially in frequency region. We suggest a method of reducing the LSP transformation time using voice characteristics The proposed method is to apply search order and interval differently according to the distribution of LSP parameters. in comparison with the conventional real root method, the proposed method results in about 46.5% reduction. And, the total computation time is reduce to about 5% in the G.723.1 vocoder.

• 대한기계학회논문집
• /
• 제15권1호
• /
• pp.28-40
• /
• 1991

본 연구에서는 4변이 단순 지지된 두꺼운 역대칭 앵글 플라이 적층판의 굽힘, 좌굴 및 진동 거동을 규명하였으며, 인장과 굽힘, 비틀림 사이의 결합 특성을 고려하 였고, 전단 변형을 고려하였다.또 유사 변환 개념을 도입하여 복합 재료 적층판의 거동을 일반화 하였다.

• EDISON SW 활용 경진대회 논문집
• /
• 제4회(2015년)
• /
• pp.246-250
• /
• 2015

In this paper, Timoshenko and Euler-Bernoulli beam theories(EB-beam) are used, and Fast Fourier Transformation(FFT) analysis is then employed to extract their natural frequencies using both analytical approach and Co-rotational plane beam(CR-beam) EDISON program. EB-beam is used to analyze a spring-mass system with a single degree of freedom. Sinusoidal force with various frequencies and constant magnitude are applied to tip of each beam. After the oscillatory tip response is observed in EB-beam, it decreases and finally converges to the so-called 'steady-state.' The decreasing rate of the tip deflection with respect to time is reduced when the forcing frequency is increased. Although the tip deflection is found to be independent of the excitation frequency, it turns out that time to reach the steady state response is dependent on the forcing frequency.

• Journal of Korean Neurosurgical Society
• /
• 제30권7호
• /
• pp.855-860
• /
• 2001

Introduction : It has been reported that the survival of low-grade glioma patients depends upon the time of malignant transformation. The authors presents the clinical analysis of histologically proven trasformed gliomas. Materials and Method : A total 92 patients who were consecutively treated and histologically confirmed hemispheric low-grade gliomas between 1980 and 1998 were analyzed and followed. All cases meet the criteria of WHO glioma classification of grade II. Results : The mean follow-up period was 73 months. Twenty two among 92 cases(24%) were histologically proven to be transformed into malignant ones. The mean time to transformation was 56 months. The 5-year and 10-year survival rates of the transformed group were 66% and 30% respectively and significantly different from the survival rates of the non-transformed group(p=0.0018). Among clinical factors at presentation, the initial tumor volume had a tendency to be larger in the transformed group than that of the non-transformed group and became significant when it was divided into more than $30cm^3$ or not(p=0.02). Among therapeutic factors, the extent of removal had no influence on the rate of malignant transformation. But postoperative radiation therapy were more frequently given to the pre-transformed group than the non-transformed group and the frequency was significantly different(p=0.02). Conclusions : The authors had found that the initial tumor volume and radiation therapy could be clinical prognostic factors for the malignant transformation of low-grade gliomas.

• 대한수학회보
• /
• 제39권4호
• /
• pp.589-606
• /
• 2002

We introduce and analyze a frequency-domain method for parabolic partial differential equations. The method is naturally parallelizable. After taking the Fourier transformation of given equations in the space-time domain into the space-frequency domain, we propose to solve an indefinite, complex elliptic problem for each frequency. Fourier inversion will then recover the solution in the space-time domain. Existence and uniqueness as well as error estimates are given. Fourier invertibility is also examined. Numerical experiments are presented.

• 한국마린엔지니어링학회:학술대회논문집
• /
• 한국마린엔지니어링학회 2002년도 춘계학술대회논문집
• /
• pp.219-223
• /
• 2002

Voltage and current signals during impulse tests on transformer are treated as non-stationary signals. A new method incorporating signal-processing method such as Wavelets and courier transform is proposed for failure identification. It is now possible to distinguish failure during impulse tests. The method is experimentally validated on a transformer winding. The wavelet transforms enables the detection of the time of occurrence of switching or failure events. After establishing the time of occurrence, the original waveform is split into two or more sections. The wavelet transform has ability to analysis the failure signal on time domain as well as frequency domain. Therefore, the wavelet transform is superior than courier transform to analysis the failure signal. In this paper, the fact was proved by real data which was achieved.

• 한국광학회지
• /
• 제19권5호
• /
• pp.370-375
• /
• 2008

본 논문에서 비동시성 광샘플링(asynchronous optical sampling; AOS) 방식을 이용하는 고속 고분해 테라헤르츠 시간영역 분광(terahertz time domain spectroscopy; THz-TDS)을 시연한다. 모터로 구동되는 선형 스테이지를 사용하지 않고, 약간 다른 반복 주파수를 갖는 두 대의 펨토초 레이저를 각각 테라헤르츠파 발생과 검출에 사용하여 고속으로 10 ns의 시간축 상의 신호를 획득하고 fast Fourier transformation(FFT)을 통하여 100 MHz의 주파수 분해능을 갖는 고분해 분광을 구현한다. Cross-correlation 방법에 의해 시간 분해능은 278 fs으로 측정되었다. 또한, 본 분광기를 이용하여 수증기의 투과 스펙트럼을 측정하고 흡수선들을 분석하였다.

• 한국정보처리학회논문지
• /
• 제2권5호
• /
• pp.691-700
• /
• 1995

프레탈을 이용한 대부분의 영상 부호화하는 최적의 유사 변환점을 얻기 위해서 방대한 비교 연산이 필요하여 장시간의 부호화 시간이 요구되는 문제점이 있다. 이러 한 문제는 블럭특성에 적합한 블럭 분류기를 설계함으로써 해결할 수 있다. 일반적으 로 공간 영역에서 보다는 주파수 영역에서 좀더 정확하고 다양한 블럭의 형태를 예측 할 수 있다. 본 논문에서는 DCT특성을 이용하여 블럭의 형태를 예측할 수 있는 블럭 분류기를 제안하였다. 이 분류기는 프랙탈 특징과 부합하여 부호와 시간을 줄임과 동 시에 복원된 영상의 화질을 높이는 장점이 있다. 주파수 영역에서의 AC 계수의 크기는 다양한 블럭 형태를 예측 가능하게 함으로써 최적의 축소 변환점을 얻기 위한 도메인 블럭과 레인지 블럭간에 비교 연산을 줄일 수 있다. 특히 DCT 각 계수의 부호값은 프 랙탈 적용을 위한 8가지 변환에 대해 불필요한 변환을 생략함으로써 2가지 변환만으로 도 축소 변환점을 찾는 장점을 나타낸다.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권10호
• /
• pp.1215-1221
• /
• 1999

A high speed digital distance relaying algorithm based on a Wavelet Transformation is proposed. To obtain stable phasor values very quickly, first, a lowpass filter which has low cutoff frequency is used. Secondly, db2(Daubechies 2) Wavelet which has the data window of 4 samples is used. A FIR filter which removes the DC-offset component in current relaying signals is applied. In accordance with a series of tests, the operation time of the relaying algorithm is less than 3/4 cycles after faults in a 80 [km], 154[kV], 60[Hz] over-head transmission line system.