• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.233-233
• /
• 2023

효율적인 수자원 관리에 필수적인 요소 중 하나는 유역 유출의 정확한 예측이다. 동일한 유역이라 할지라도 과거 기후조건에 대해 매개변수나 모형구조가 최적화된 수문모형은 현재나 미래 기후에 대해 최적이라 할수 없으며, 이에 따라 유역 유출 해석의 불확실성 또한 증가하고 있다. 수문자료동화는 모형의 입력 자료에 따른 불확실성을 줄이고 예측정확도를 향상 시킬 수 있는 방법으로, 수문모형의 상태량이나 매개변수를 업데이트하여 모형 초기 조건의 가능성 높은 추정치를 생성하는 기법이다. 본 연구에서는 국내 댐 상류 유역에 대해 집중형 수문모형과 순차자료동화 기법의 연계 패키지인 airGRdatassim 모형을 적용하여, 앙상블 칼만 필터와 파티클 필터 기법의 수문자료동화 특성을 비교 분석하고, 자료동화와 관련된 하이퍼-매개변수의 불확실성이 수문모의 성능에 미치는 영향을 분석하였다. 자료동화 적용 결과, 두 자료동화 기법 중 파티클 필터에 의한 모의성능이 높았으며 기상강제력 노이즈의 범위, 갱신 대상 상태량 설정, 앙상블 설정 등 수문자료동화의 설정과 관련된 하이퍼 매개변수의 불확실성은 두 기법별 뚜렷한 차이를 보였다. 또한, 본 연구에서는 일단위에서 시단위로 확장한 유량 예측 자료동화의 시험 모의결과 및 앙상블 수문동화기법의 도전과제에 대해서도 논의한다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.10
• /
• pp.181-186
• /
• 2012

In this paper, we describe the synthesis of robust and non-fragile Kalman filter design for a class of uncertain linear system with polytopic uncertainties and filter gain variations. The sufficient condition of filter existence, the design method of robust non-fragile filter, and the measure of non-fragility in filter are presented via LMIs(Linear Matrix Inequality) technique. And the obtained sufficient condition can be represented as PLMIs(parameterized linear matrix inequalities) that is, coefficients of LMIs are functions of a parameter confined to a compact set. Since PLMIs generate infinite LMIs, we use relaxation technique, find the finite solution for robust non-fragile filter, and show that the resulting filter guarantees the asymptotic stability with parameter uncertainties and filter fragility. Finally, a numerical example will be shown.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.5
• /
• pp.549-554
• /
• 2012

In this paper, we consider the optimal fuzzy filter of nonlinear discrete-time with estimation error variance constraint. First, the Takagi and Sugeno(T-S) fuzzy model is employed to approximate the nonlinear system. Next, the error state is mean square bounded, and the steady state variance of the estimation error of each state is not more than the individual predefined value. It is shown that, the addressed problem can be carried out by solving linear matrix inequality(LMI) and some algebraic quadratic matrix inequalities. Finally, some examples are provided to illustrate the design procedure and expected performance through simulations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.12 no.5
• /
• pp.466-477
• /
• 1987

본 논문에서는 NTSC칼라 TV합성신호를 프레임내에서 색도신호와 명도신호성분으로 분리하기 위한 두 종류의 적응 필터링 방식을 제시하였다. 적응 필터링 방식에 있어서 합성신호는 수직필터와 수평필터에 의하여 필터링 되어 지고 화상의 국부적인 특성에 따라 필터의 출력이 선택되어 진다. 첫번째 방식에서는 조건형 스위칭 알고리즘에 의하여 수직필터 또는 수평필터의 출력이 최종단의 출력으로 결정되어 진다. 두번째 방식에서는 선형 조합 검출 알고리즘에 의하여 수직필터와 수평필터의 출력의 가중치합이 최종단의 출력이 된다. 사용된 필터들은 NTSC 칼라 TV 합성신호를 4fsc로 샘플한 경우에 대하여 설계되어졌다. 몇가지 정량적인 기준을 이용하여 여러가지 방식들을 컴퓨터 시뮬레이션에 의하여 비교평가하였다.

• Korean Journal of Optics and Photonics
• /
• v.19 no.1
• /
• pp.36-42
• /
• 2008

Implementation of a wavelength-swept source with constant tuning rate adopting a PZT-type tunable filter, requires the knowledge of hysteresis of the filter used. The hysteresis must be considered to avoid any degradation in resolution of the optical frequency domain reflectometry (OFDR) system. An optical spectrum analyzer (OSA) could be used to do the hysteresis measurement, but its measurement time is too long for the high-speed driving conditions for the filter. We proposed a new hysteresis measurement method based on OFDR, which could measure the hysteresis in a real driving condition. A hysteresis measurement apparatus consisted of wavelength-swept source, interferometer, signal processing unit, and PC program is built and used to do the measurement. It is concluded that the new method is useful in the measurement of hysteresis at real driving conditions by successfully implementing a swept-wavelength source whose wavelength change is linear in time.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.285-294
• /
• 2004

Median Filters in the Signal Processing have been most widely used and have demonstrated the most strongest effects. This paper proposes the Adaptive Median Filters by using noise detection. The basic algorithm of the proposed filters is to determine whether noise or not by the each noise judgement standards, and then take the Median Filter if it satisfies the conditions as a result of judgement and returns to the original image(No Filters) if not. This paper presented Noise Detection by Local Variance and Local Central Variance for noise judgement, compared and analyzed the features and performance of existing [5]∼[10] Filters. Filter improved on the result of executing the existing filters at the same condition and showed the effects over that when it was judged with naked eyes. Accordingly, the Adaptive Median Filters by Local Variance and Local Central Variance was proven to have reinforced edge preservation ability and have the strong features for removing the Impulse Noise of the Median Filter.

• Proceedings of the KIEE Conference
• /
• 2006.07b
• /
• pp.1020-1021
• /
• 2006

전력용 능동 필터를 제어하기 위해서는 보상 고조파 전류에 대한 지령치가 필요하다. 3상 전력용 능동 필터의 경우 순시전력이론을 이용하여 보상 전류 지령치를 계산할 수 있다. 하지만 단상 전력용 능동 필터의 경우에는 디지털 필터를 사용하는 것이 일반적이다. 디지털 필터의 차수가 높은 경우에는 필터의 성능이 향상되지만 높은 차수에 따르는 연산시간의 증가와 민감도 증가로 인하여 설계 조건에 부합하는 저차 디지털 필터의 설계가 필요하다. 본 논문은 제안된 notch filter 설계 기법으로 디지털 필터를 설계하고 제어루프에 필터 사용시 발생 할 수 있는 안정도 저하를 평가하고, 시뮬레이션을 통하여 그 성능을 검증하였다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.4
• /
• pp.56-64
• /
• 2000

The design of filters for pattern recognition that have optimal trade-off for the criteria of noise robustness, peak sharpness and Homer efficiency, when input scene noise is spatially disjoint (nonoverlapping) with the target, are presented Three different criteria, ie, signal-to-noise ratio, peak-to-output energy ratio and light efficiency are used to design the filter. Computer simulation is made for the various types of noise and parameters to illustrate filter performance for optical pattern recognition. When we compare the new trade-off filter with the original optimal trade-off filter, the performance of trade-off filter for nonoverlapping background is better than that of trade-off filter for overlapping background.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.6
• /
• pp.539-547
• /
• 1992

Considering the field of high frequency or broad-band signal filtering with high speed, we pro-pose optical fiber filters which consist of directional couplers and optical fiber delay elements. Fiber delay line signal processor Is operated with Incoherent optical carrier because of Its simplicity. Due to the characteristics of the directional coupler which Is basic element of optical fiber filter a new design method which Is different from that of usual digital filter is required. By considering the characteristics of directional coupler and in order to make good use of the optical signal energy, we have given the design formulae and the conditions of optical fiber filters for the direct and the lattice form.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.4
• /
• pp.1051-1058
• /
• 1998

We propose a bandpass filter using wavelength selectivity as variation of writing beam incident angle in photorefractive crystal. From TWM(Two-Wave Mixing) experiment, optimal diffraction properties are induced, and bandpass filter for selection wavelength transmission bandwidth are taken by change of writing beam incident angle.