• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.65-66
• /
• 2007

본 논문은 파라미터 변동에 잘 적응하며 강인한 영구자석 동기전동기의 위치센서 없는 센서리스 제어를 위해 개선된 슬라이딩 모드 관측기에 고정자 저항 추정기능을 추가한 새로운 슬라이딩 모드 적응 관측기를 제안한다. 제안된 관측기는 개선된 슬라이딩 모드 관측기를 사용함으로써 기존의 슬라이딩 모드 관측기에서 채터링(Chattering) 문제 해결을 위해 저역통과필터(Low-pass filter)의 사용 및 추가적인 회전자의 위치보상은 Sigmoid 함수를 스위칭 함수로 사용하므로 제거 할 수가 있다. 또한 고정자 저항 추정기능을 추가하여 센서리스 운행 중 환경적 요인으로 인한 고정자 저항값의 변동이나 초기에 고정자 저항값을 잘못 알고 있더라도 빠른 적응을 통해서 추정오차를 절감시켜 전동기의 속도추정 성능을 향상시킬수 있다. 또한 기존의 적응 슬라이딩 모드 관측기에서의 적분연산을 줄임으로써 제어시스템의 성능을 개선시켰다 제안된 관측기의 안정성은 Lyapunov 후보 함수를 이용하여 관측기의 이득을 설정함으로써 검증하였으며 관측기의 성능은 시뮬레이션을 통한 실험을 통하여 그 타당성을 입증하였다.

• Composites Research
• /
• v.28 no.6
• /
• pp.383-388
• /
• 2015

Ionic Polymer-Metal Composite (IPMC) consisting of soft membrane plated by platinum electrode layers on both surfaces generates electric energy when subjected to various mechanical stimuli. The paper proposes a circuit model that describes the physical composition of IPMC to predict the voltage generation characteristic corresponding to bending motion. The parameter values in the model are identified to minimize the RMS error between the real and simulated outputs. Following the design of IPMC circuit model, the state observer of the model is designed by using pole placement technique which improves the model accuracy. State observer design technique is also applied to find the inverse model which estimates the input bending angles from the output voltage data. The results show that the inverse model estimates input bending angles fairly well enough for the further applications of IPMC not only as an energy harvester but also as a bending sensor.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2
• /
• pp.40-49
• /
• 1995

Application of the filtered-x LMS adaptive filter to active noise cancellation requires to estimate the transfer charactersitics between the output and the error signal of the adaptive canceler. In this paper, we derive the filtered-x adaptive noise cancellation algorithm and analyze its convergence behavior when the acoustic noise consists of multiple sinusoids. The results of the convergence analysis of the filtered-x LMS algorithm indicate that the effects of the parameter estimation inaccuracy on the convergence behavior of the algorithm are characterized by two distinct components : Phase estimation error and estimated gain. In particular, the convergence is shown to strongly affected by the accuracy of the phase response estimate. Simulation results are presented to support the theoretical convergence analysis.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.27 no.1
• /
• pp.103-108
• /
• 2023

We study the problem of selecting a subset of sensor nodes in sensor networks in order to maximize the performance of parameter estimation. To achieve a low-complexity sensor selection algorithm, we propose a greedy iterative algorithm that allows us to select one sensor node at a time so as to maximize the log-determinant of the inverse of the estimation error covariance matrix without resort to direct minimization of the estimation error. We apply QR factorization to the observation matrix in the log-determinant to derive an analytic selection rule which enables a fast selection of the next node at each iteration. We conduct the extensive experiments to show that the proposed algorithm offers a competitive performance in terms of estimation performance and complexity as compared with previous sensor selection techniques and provides a practical solution to the selection problem for various network applications.

• Geophysics and Geophysical Exploration
• /
• v.10 no.3
• /
• pp.191-202
• /
• 2007

To identify the detailed attenuation structure in the southern Korean Peninsula, a numerical test was conducted for the Q tomography inversion to be applied to the accumulated dataset until 2005. In particular, the stochastic pointsource ground-motion model (STGM model; Boore, 2003) was adopted for the 2D Q tomography inversion for direct application to simulating the strong ground-motion. Simultaneous inversion of the STGM model parameters with a regional single Q model was performed to evaluate the source and site effects which were necessary to generate an artificial dataset for the numerical test. The artificial dataset consists of simulated Fourier spectra that resemble the real data in the magnitude-distance-frequency-error distribution except replacement of the regional single Q model with a checkerboard type of high and low values of laterally varying Q models. The total number of Q blocks used for the checkerboard test was 75 (grid size of $35{\times}44km^2$ for Q blocks); Q functional form of $Q_0f^{\eta}$ ($Q_0$=100 or 500, 0.0 < ${\eta}$ < 1.0) was assigned to each Q block for the checkerboard test. The checkerboard test has been implemented in three steps. At the first step, the initial values of Q-values for 75 blocks were estimated. At the second step, the site amplification function was estimated by using the initial guess of A(f) which is the mean site amplification functions (Yun and Suh, 2007) for the site class. The last step is to invert the tomographic Q-values of 75 blocks based on the results of the first and second steps. As a result of the checkerboard test, it was demonstrated that Q-values could be robustly estimated by using the 2D Q tomography inversion method even in the presence of perturbed source and site effects from the true input model.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.36 no.5
• /
• pp.455-461
• /
• 2008

In this paper, the performance of a scalar filter and a scalar adaptive filter are analyzed. In order to make indoor experimental environment similar to outdoor test, ultrasonic sensors are used instead of GPS. The scalar adaptive filter, which is continuously estimating velocity error covariance and measurement noise covariance by using adaptive method, is different from the scalar filter. Experimental results show that the scalar adaptive filter has better position estimating performance than the scalar filter by estimating above two parameters with an adaptive method.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.4
• /
• pp.336-342
• /
• 2009

A conventional Closest Point of Approach (CPA) analysis allows a non-maneuvering moving source that is radiating a constant frequency tone to be located using doppler shifted frequency measurements obtained by a stationary receiver. The original frequency, relative speed of the target, time at the CPA, and range from the CPA to the sensor are estimated by the conventional CPA. However, this paper proposes a new CPA analysis that allows the motion parameters of a target to be estimated using the bearing and frequency measurements obtained by a moving receiver that has a constant velocity. The validity of the proposed estimation scheme is confirmed through a performance analysis and simulation study.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.16 no.1
• /
• pp.37-45
• /
• 2012

In an effort to further exploit the peak ground-motion acceleration (PGA) information per second available in real time by the enacted law, bracketed summations of the PGA per second ($BSPGA_k$) for 30 seconds based on the records with a rate of 100 samples were compared with the cumulative absolute velocity (CAV) and earthquake intensities based on a worldwide database of records from small-to-large earthquakes. The CAV, currently in use as an earthquake damage indicator for nuclear power plants due to its strong correlation with the earthquake intensity, has the disadvantage of requiring a massive amount of digital data with a rate of more than 100 samples per second. The comparative study shows that the $BSPGA_k$ is well correlated with the CAV over the wide range of strong ground-motion levels, which suggests that the $BSPGA_k$ is one of the new promising ground-motion parameters especially useful for rapid earthquake alert notifications through an earthquake monitoring network. Based on the domestic database of records from small-to-moderate earthquakes with felt reports, it is also observed that the $BSPGA_k$ is comparable to the CAV and better than the PGA in predicting the intensity by using the correlation relation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.11a
• /
• pp.225-228
• /
• 2011

동영상을 고효율로 압축할 수 있는 중요한 기술 중의 하나가 움직임 추정 기술이다. 움직임 추정 기술에서 가장 널리 사용되는 탐색 방식과 정합오차를 판단하는 기준은 각각 전역 탐색(FS: Full Search)알고리듬과 Sum of Absolute Differences(SAD)이다. 하지만, FS와 SAD를 사용한 움직임 추정기술은 인코더 전체 계산량의 50%이상을 차지하게 된다. 무선 통신 및 인터넷 환경에서 사용되는 응용프로그램에 대해서는 많은 계산량이 실시간 전송을 어렵게 만든다. 이러한 이유로, 동영상 압축 기술에서 계산량을 줄이는 방법이 중요하게 되었다. 그 결과로, SAD대신 다른 정합 오차 기준을 사용하여 움직임 추정기술의 계산량을 줄이는 알고리듬들이 제안되었다. 본 논문에서는 계산량을 줄이면서 하드웨어 구현 시 많은 이점을 가진 1비트 변환 (One-bit Transform, 1BT)알고리듬과 1비트 변환의 성능을 향상시킨 제한된 1비트 변환(Constrained one-bit transform, C-1BT)을 소개하고 C-1BT방법에 있어서 기존 방식의 빠른 인코딩속도를 유지하면서 PSNR을 증가시키는 알고리듬을 제안한다. 제안하는 알고리듬에서는 기존 C-1BT에서 고정해서 사용한 파라미터 D값을 대신에 현재블록과 동일한 위치에 있는 참조영역의 블록과 그 주변블록의 움직임벡터 정보를 이용해서 블록의 움직임 정도를 판단하고, 블록단위로 D값을 적응적으로 결정한다. 실험결과는 제안하는 알고리듬이 기존의 알고리듬과 비교하여 평균 0.11dB PSNR 증가를 보여 준다.

• Journal of Ocean Engineering and Technology
• /
• v.13 no.3B
• /
• pp.83-90
• /
• 1999

본 논문에서는 금후 널리 사용되리라고 예상되는 장대 유연한 해양구조계의 동적 응답에 대한 능동제어를 다루었다. 제어기법으로는 optimal, adaptive 제어와 우주구조물의 진동제어에 사용되는 LAC/HAC 제어를 도입하여 simulation과 모형실험을 수행하였다. 기존의 발표된 최적제어 실험결과는 만족스럽지 못하여 다시 수행할 필요성이 있었다. 그 결과, 해양구조계의 위치 및 탄성변형이 타당한 제어력 범위내에서 제어가 되어 본 연구에서 제안한 제어시스템의 유효성이 입증되었다. 또한 제어기법간의 비교를 통하여 장대 유연한 해양구조계에 대한 최적의 제어기법에 대하여 조사를 하였는데, 초기 시스템 파라미터의 추정에 오차가 있는 경우 적응제어 알고리즘의 장점을 확인할 수가 있었다.